Nextgov/FCW - Defense

Ready, fire, aim: Pentagon cut workforce with little analysis before or since, GAO finds

Meghann Myers — Mon, 01 Jun 2026 18:00:00 -0400

Pentagon leaders cut their department’s workforce by more than 10 percent with little regard for the effects—and still has no plans to assess them, according to a congressional watchdog report released on Friday.

The department shed 78,000 civilian employees in 2025 through a mix of voluntary resignations, involuntary layoffs, and a hiring freeze that resulted in nearly 60,000 fewer new hires than in recent years, the report found.

“But we found that DOD didn’t consistently analyze the impacts of these reductions, either in 2025 or in prior years,” according to the report. “DOD also doesn’t have a plan to assess lessons learned from its 2025 workforce reductions.”

In their response to the report, Defense officials agreed that they should “develop and implement a plan for collecting and sharing lessons learned from the Department's implementation of workforce reduction efforts.”

The officials did not indicate whether that would happen.

Soon after Defense Secretary Pete Hegseth took office, the Pentagon announced it would cut 5 to 8 percent of its civilian workforce. Within a year, the number swelled to about 110,000—about 14 percent of DOD civilians—including laid-off probationary employees, deferred resignations, and voluntary early retirements. Some 30,000 people were hired for a short list of jobs exempted from the hiring freeze, putting the net loss at just over 10 percent.

Of the 28 Defense agencies, offices, and other organizations targeted for workforce cuts by the Trump administration’s fiscal 2026 budget request, at least three did not give the required explanation to Congress about why and how the cuts would be made, GAO found.

Those were the Joint Staff, Defense Threat Reduction Agency, and the Defense Contract Audit Agency, according to the report.

“According to component officials, DOD had not provided guidance for when and how to conduct and document this analysis,” the GAO found.

And further, the GAO found, the Pentagon didn’t plan to assess how the cuts affected productivity.

In March, the Partnership for Public Service published a survey that found morale among DOD employees has tanked during the current administration.

Only 9 percent of Army Department employees agreed that “Secretary of War Pete Hegseth’s political leadership team generates high levels of motivation in the workforce,” the survey found, the most satisfied of any of the large government agencies surveyed.

]]>

ODNI assigns two officials to lead intelligence coordination on election threats

David DiMolfetta — Thu, 14 May 2026 18:21:00 -0400

The Office of the Director of National Intelligence recently named two officials to a role coordinating with the nation’s spy agencies on threats against the 2026 midterm elections, according to a congressional source and a second person familiar with the matter.

Dave Mastro and James Cangialosi will jointly oversee the intelligence community’s election threat mission, serving in the role of election threats executive. Both sources requested anonymity to communicate the appointments.

Mastro serves on the National Intelligence Council, which produces intelligence assessments drawn from findings across the nation’s spy agencies, including reports requested by Congress and senior policymakers. Cangialosi serves as deputy director of the National Counterintelligence and Security Center.

“We have an expansive team of professionals at ODNI focused on carrying out President [Donald] Trump’s and [Director of National Intelligence Tulsi] Gabbard’s election integrity efforts,” which includes Mastro and Cangialosi, ODNI spokesperson Olivia Coleman said in a statement.

The office is also “providing robust briefings, on par with efforts traditionally carried out during election years, to protect election integrity this midterm cycle,” Coleman said.

For months, it was unclear if ODNI ever named an election threats executive responsible for leading the intelligence community on election security for the coming midterm cycle.

The Record first reported the appointments.

Established in 2022, the Foreign Malign Influence Center was designed to coordinate spy agencies’ efforts to identify and assess foreign influence and disinformation threats targeting elections. But an overhaul inside ODNI launched last summer shifted many of the center’s responsibilities to the National Counterintelligence and Security Center and the National Intelligence Council, with ODNI arguing the previous structure raised constitutional concerns over coordination with social media companies.

The election threats executive — created in 2019 during Trump’s first term — typically oversees an “Experts Group” that analyzes intelligence on foreign interference efforts.

Election threats can include cyberattacks on voting systems, foreign influence operations and disinformation campaigns aimed at undermining public trust in elections.

The assignments come as Gabbard has faced criticism over her involvement in the White House’s broader review of election security outcomes, including scrutiny from Democrats tied to her presence during an FBI raid on a Georgia election office and ODNI-led examinations of voting machines in Puerto Rico.

This year’s annual intelligence assessment of worldwide threats to the U.S. did not describe foreign threats to the nation’s elections, the first time in nearly a decade.

Trump has continued to falsely claim the 2020 election was stolen from him, despite courts, audits and state reviews finding no evidence of widespread fraud that would have changed the outcome.

The appointments also come amid broader changes to the federal government’s election security apparatus ahead of the 2026 midterms. In recent months, Democrats and state election officials have raised concerns over cuts to election-focused programs at the Cybersecurity and Infrastructure Security Agency, which has lost around a third of its workforce in the last year.

]]>

SOCOM adding AI, autonomy ‘at every level,’ commander says

Patrick Tucker — Wed, 29 Apr 2026 12:00:00 -0400

AI and autonomy are being integrated into special operations “at every level,” the leader of U.S. Special Operations Command told lawmakers on Tuesday—an indication that SOCOM, like smaller organizations everywhere, is well-poised to take advantage of disruptive technologies.

They are “critical” to sensing the battlefield, continuously surveilling adversary forces and targets, and “the ability to project violence, should that be required,” Adm. Frank “Mitch” Bradley said at a Senate Armed Services Committee hearing. He added that they are also key to improving international partners, underscoring their particular value to special operations.

Bradley’s testimony underscored a larger phenomenon playing out in boardrooms as well as on battlefields: small and nimble groups—whether non-state actors, software startups, or militaries like Ukraine’s—derive greater return on their AI investments than do more established or incumbent players. The boom in market valuation of small AI-focused defense startups like Anduril, Shield AI, or Swarmer versus slower growth of traditional players tells that story, as does Ukraine’s use of drones and autonomy to withstand Russia’s invasion.

SOCOM is better positioned to adopt AI than, say, the U.S. Navy, which is also trying to fund and sustain multibillion-dollar aircraft carriers and other maintenance-thirsty warships. Even the Navy’s forays into autonomy tend to be on the larger side, like its plan to spend $6 billion to acquire 70 Medium Unmanned Surface Vessels. At a hearing last week during a hearing, Sen. Rick Scott, R-Fla., responded with clear disapproval at spending that amount of money to get 70 surface craft—good value compared to the cost of a destroyer but not compared to the Ukrainian robot boats that have corralled the Russian Navy.

Meanwhile, SOCOM has launched a broad request to industry for new ideas for maritime autonomy, human performance, command-and-control technology, and “scalable effects”—as in technologies that can be increased in number or intensity, such as directed energy, electronic warfare, cyber-enabled effects, and precision engagement tools. The request bespeaks flexibility in a way that the Navy’s 70-MUSV order does not.

AI enables asymmetric warfare—SOCOM’s specialty—more than traditional warfare, and the special operations command has fewer obstacles than the service branches to fast implementation. A case in point is Maven, an Air Force Special Operations Command tool for video and data analysis that has become a widely used program of record

“From the battlefield to the back office,” SOCOM is ‘finding ways to be able to bring autonomy, attritable, mass autonomy, to bear is a very important part of how we on the edge can leverage our placement and access,” Bradley said.

Partners

But SOCOM also plays a key role in helping partner militaries develop new capabilities quickly. Bradley described how SOCOM is seeking to use AI and autonomy “not only to serve our own interests, but to be able to help our partners who generally don't have the same budgets we do, to be able to buy that kind of capacity to give them asymmetric advantages… I think that's critical, because it is not just about what we bring, but it's about enabling those partners.”

Bradley, similar to other military leaders, pointed out the unique relationship that the U.S. military has forged with Ukraine. Special operations forces were critical to helping Ukraine stand up new concepts and tactics to thwart Russia’s advance in 2022. Today, that relationship provides key knowledge and training benefits back to U.S. Special Operations Forces. “Frankly, we learn from them,” he said.

That partnership is especially critical in finding real-world and relevant data to inform SOCOM training, concepts, and buying, namely testing new gear under spectrum-warfare conditions and real-world threats. Meaningfully testing new equipment for use against an actual modern adversary “requires more exquisite ranges that have the ability for us to be able to practice, train and rehearse in increasingly contested electromagnetic spectrum environments. Those are difficult to be able to produce inside the United States,” he said. “We have to be able to bring together our standard, exquisite weapons systems now with teamed and collaborative autonomy, and there are very few places inside the United States where that is an easy thing to do, in many places where it needs to be done, and we are working to do that.”

Industry, however, has the most to gain from longstanding military-to-military partnerships with Ukraine and its innovative forces and companies.

“I will say that many of our business, our defense industrial base partners, are watching this as well,” he said. “And of course, the Ukrainians are driven by the existential need for that cycle of adaptation. As we watch that, I have great confidence that our industrial base here can do the same.”

]]>

Pentagon adds Google’s latest model to GenAI.mil as usage soars

Frank Konkel — Mon, 27 Apr 2026 12:00:00 -0400

LAS VEGAS — Users of the Pentagon’s enterprise-wide generative-AI platform now have access to Google Cloud’s latest and most advanced commercial AI model, Gemini 3.1 Pro, after several weeks of using the software in preview mode.

The software is available to defense users through the GenAI.mil platform and will also be available for all Gemini for Government users across the federal government.

“Gemini 3.1 Pro is Google’s most sophisticated model yet, and it really represents the frontier of American AI,” Pentagon Chief Data Officer Gavin Kliger said in an interview Thursday. “And so the department is working with our engineering team together to make sure we can have this capability available across the department.”

It’s the latest upgrade for the platform, which launched in December with initial plans to integrate Gemini for Government, and later announced plans to incorporate AI models from OpenAI and xAI.

Up to 3 million users have access to GenAI.mil, which is actively being used by more than 1.3 million of them, Kliger said. They are tapping the generative-AI software to automate tasks and workloads, streamline laborious processes, and disseminate and summarize data-heavy documentation in the department’s Impact Level 5 environments, which handle sensitive unclassified data.

Kliger’s team offered several examples of how defense personnel are using generative AI shared by the department. One user at Navy Recruiting Command used Gemini to cut the time to build an automated database to manage personnel and accounts from several years to three months, saving an estimated 10 weeks of labor annually. And a lab director at the Defense Logistics Agency used generative AI to reduce the time to draft statements of work “from weeks to hours,” helping secure $1 million in last-minute funding for a laboratory modernization.

Google Public Sector Chief Executive Officer Karen Dahut said the company’s decision to software-define its commercial cloud and bring it to the government market—as opposed to using physically separate data centers for government users—allows the company to accredit its commercial software faster than competitors. That speed-to-market advantage is most critical in defense missions, she said.

“The key thing is that this [Gemini 3.1 Pro] is our most recent, newest, most capable model, period,” Dahut said on the sidelines of Google Cloud Next in Las Vegas. “So GenAI.mil users are getting it only eight weeks behind all the commercial customers,” who gained access in February.

GenAI.mil, launched on Dec. 9, accumulated 500,000 users within a week and 1 million users within a month “with zero latency issues and zero downtime,” Dahut said.

“No other cloud provider could have launched at that scale,” she said.

Kliger said “one of the cool things about all the growth is that it’s been organic”: the department makes modern AI tools available to its employees but isn’t prescriptive about their use.

Kliger also said increased collaboration with industry is critical in ensuring AI dominance over adversaries including China.“The work we do now is going to set the tone for the next decade, and these are super important technologies for the national defense, our national security,” Kliger said. “China, of course, has a really tight collaboration between the government and its private sector, effectively a controlling relationship. And so making sure we’re engaging with the frontier labs, working together closely like we are with Google, is incredibly important for the nation. Google has been a great partner with the department.”

The Pentagon enters its vibe-coding era

Over the past several weeks, users have tapped another Google Cloud product made available through GenAI.mil, Agent Designer, to vibe-code thousands of agentic AI agents. AI agents are autonomous systems that use large language models such as Gemini to perform tasks without human intervention at each turn.

At the Box Federal Summit Thursday, Jacob Glassman, deputy assistant defense secretary for science and technology foundations in the research and engineering directorate, said users had already built more than 100,000 AI agents on GenAI.mil.

These AI agents have authorizations to operate at IL5, which means they can be used for the department’s most sensitive unclassified data. They also don’t require much coding experience or training to use.

Agent Designer “allows anyone, be they technical or not, to kind of use natural language to describe the system they want to set up,” Kliger said. “One of the big changes we’re seeing is moving from the old concept of the large language models being just a chat interface to being an actual platform where it can run tasks on its own.”

]]>

Autonomous weapons will be ‘key and essential part’ of warfare, Joint Chiefs chair says

David DiMolfetta — Thu, 23 Apr 2026 13:35:00 -0400

NASHVILLE — Joint Chiefs Chairman Gen. Dan Caine said Thursday that autonomous weapons are going to be a “key and essential part of everything we do” when asked about how such tools would fit into the future of warfare.

Speaking during a fireside chat at Vanderbilt University’s Asness Summit on Modern Conflict and Emerging Threats, Caine said “we are doing a lot of thinking about this in the joint force right now” on how autonomous tech would be applied to areas like drones and command-and-control operations.

His remarks signal that the U.S. military is keen on crafting plans to further adopt artificial intelligence tools and other evolving technologies that would automate national security decisions made in the Defense Department.

“Probably everybody in this room uses some flavor of a [large language model] every single day,” he said, adding the same can’t be said for staff in the halls of the Pentagon. “So, we have to really normalize this and become early adopters.”

The remarks come as observers weigh tensions between the Pentagon and Anthropic, which recently unveiled a powerful frontier AI model, Mythos Preview, that was held back from public release over cybersecurity risks, paired with a new initiative to study its effects on global networks.

Intelligence community units have expressed interest in Mythos, Nextgov/FCW previously reported. The NSA, a component of the DOD, has been granted access to it, Axios reported Sunday.

Earlier this year, Anthropic declined to ease restrictions against its tools being used for domestic surveillance or fully autonomous weapons for Pentagon use, triggering a “supply chain risk” designation from the Defense Department and a White House order that all federal agencies phase out their uses of Anthropic tools.

The company has legally challenged the move, and a federal judge issued a temporary injunction on the designation and ban in late March. The government has said it intends to appeal the injunction.

This week, President Donald Trump said in a CNBC interview that the company is “shaping up” and can “be of great use” in the future, a sign that tensions between Anthropic and the government may be easing up.

The use of AI in military operations often draws scrutiny because it can speed up battlefield decisions while blurring human accountability, and it can raise doubts about whether such systems would reliably comply with the laws of war. Lawmakers have asked the Pentagon if AI systems were used in a deadly strike on a school in Iran that occurred in the opening hours of the U.S.-Israel war against Tehran.

Caine also said U.S. government agencies need to be “better buyers” for the private sector. “We have to write better contracts,” he said, elaborating that current acquisition frameworks are slowing contract workflows.

Contracts should be structured so risk is shared between buyers and sellers with the goal of bringing better outcomes for servicemembers, he added.

]]>

CIA deception campaign helped US rescue downed airman in Iran, director says

David DiMolfetta — Mon, 06 Apr 2026 14:32:00 -0400

A deception campaign launched by the CIA bought time for U.S. forces to rescue an airman who went down in Iran on Friday, CIA Director John Ratcliffe said in a White House news conference on Monday.

The CIA deployed human assets and “exquisite technologies” to contribute to the rescue of the weapons systems officer of an F-15E Strike Eagle, Ratcliffe said. The aircraft’s pilot was rescued earlier upon the crash, but Iran was “desperately hunting” for the backseater who ejected further from his wingman and had moved away from the crash site.

The injured officer was found by the CIA in a mountain crevice but was still invisible to Iranian forces.

“Following the successful exfiltration on Saturday night, our intelligence reflects that the Iranians were embarrassed and ultimately humiliated by the success of this audacious rescue,” Ratcliffe said.

At the press conference, President Donald Trump said it was the CIA’s “genius” that contributed to the rescue and that the spy agency had spotted “something moving up the mountain.”

“This is at night. And they kept the camera on him for 45 minutes,” Trump said, suggesting the CIA had a covert surveillance capability — potentially a drone or satellite — available to track the airman’s movements.

The public remarks about the mission show how the Trump administration has made it a point to highlight contributions that CIA operatives have made toward its national security efforts.

Those include operations that targeted the government of ousted Venezuelan leader Nicolás Maduro. The agency has also taken a more public-facing posture, releasing recruitment videos aimed at sourcing in China. And in the months leading up to the Iran war, agency spies had been reportedly tracking the movements of now deceased Ayatollah Ali Khamenei.

]]>

Annual intelligence assessment doesn’t address foreign threats to US elections

David DiMolfetta — Wed, 18 Mar 2026 15:06:00 -0400

An annual intelligence assessment of worldwide threats to the U.S. omitted mentions of foreign threats to American elections for the first time in nearly a decade, a notable shift in a midterm election year that suggests the Trump administration is shifting focus away from a risk long treated as central to national security.

The assessment was delivered on the heels of a major global threats hearing in the Senate Intelligence Committee on Wednesday, where top officials including Director of National Intelligence Tulsi Gabbard and CIA Director John Ratcliffe testified about the ongoing Iran war and other top-of-mind matters.

The hearing highlighted growing tensions between intelligence assessments and the administration’s framing of the conflict with Tehran. It also came a day after the high-profile departure of Gabbard aide and National Counterterrorism Center Director Joe Kent, who said he could not agree with the Trump administration’s premise for the Iran war that began Feb. 28.

Gabbard drew ire from committee Democrats over election threats matters.

“Are you saying there is no foreign threat to our ⁠elections in the midterms this year?” Virginia Sen. Mark Warner, the panel’s top Democrat, asked Gabbard.

"The intelligence community has been and continues to remain focused on any collection and intelligence that show a potential foreign threat,” she said.

Gabbard has drawn scrutiny over her involvement in an FBI raid of a Fulton County, Georgia elections office that was at the center of President Donald Trump’s false claims of election fraud in 2020. Gabbard’s agency, in part, is charged with countering foreign election interference, and doesn’t have conventional authority to manage domestic election affairs.

Asked about this, she said the Office of the Director of National Intelligence has “purview and overview” over the Department of Homeland Security and the FBI, both of which “have purview over election security responsibilities to ensure the integrity of our elections.”

Gabbard said she only observed the raid and that she “did not participate in a law enforcement activity, nor would I, because that does not exist within my authorities.”

Gabbard’s election integrity efforts have involved multiple agencies and senior officials, including meetings this year with White House Chief of Staff Susie Wiles and then-DHS Secretary Kristi Noem to discuss election security and restoring public trust, a U.S. official previously told Nextgov/FCW. The discussions have also included outside figures like Kurt Olsen and Cleta Mitchell, both of whom have promoted claims that the 2020 election was stolen.

Warner also criticized intelligence agencies for not responding to committee requests for briefings regarding foreign election interference efforts.

Ire over Iran war

Gabbard said in her opening remarks on the Iran war that Tehran could face mounting pressure as its economy weakens, but warned that the country and its proxies “continue to attack U.S. and allied interests in the Middle East” despite setbacks before and after the conflict began.

But she notably deviated from her prepared remarks to the Senate panel and said Iran was “trying to recover from the severe damage to its nuclear infrastructure sustained during the 12-Day War” last summer, which concluded in the U.S. Midnight Hammer operation that targeted three key Iranian nuclear enrichment sites.

Per her written statement, she was expected to say Iran made “no efforts” since the U.S. bombing of their nuclear facilities “to try to rebuild their enrichment capability.” Those remarks could undermine the Trump administration’s motives to justify an attack on Iran, on the grounds that its nuclear program still posed a threat, among other reasons.

Warner asked Gabbard why her testimony diverged from her prepared remarks. She said she skipped some portions because “time was running long” during her opening statement, prompting Warner to accuse her of omitting “the parts that contradict the president.”

The state of Iran’s nuclear capabilities have especially been a flashpoint since the Midnight Hammer bombing last summer. A preliminary Defense Intelligence Agency assessment appeared to undercut Trump’s claims that Iran’s nuclear program was “obliterated” in those attacks, though the CIA soon after said it had evidence proving the program was severely damaged.

Ratcliffe told senators Wednesday that Midnight Hammer was successful and has slowed Iran’s nuclear enrichment efforts.

“We sit here today with Iran having exactly the same amount of enriched uranium to 60%, meaning they have been unwilling and uncapable, or incapable, of enriching uranium to 60%” as a result of the operation, he told Republican Sen. Mike Rounds of South Dakota.

Questions were also raised about other foreign adversaries sharing intelligence with Iran to target U.S. forces in the Middle East. Iran is “requesting intelligence assistance from Russia, from China and from other adversaries of the United States,” Ratcliffe told Sen. Jack Reed, D-R.I., declining to mention in public session if they actually are providing it. He said he knew the answer and would explain in a classified session.

Ratcliffe told Sen. John Cornyn, R-Texas, that he disagreed with Joe Kent’s claims about Iran, saying “intelligence reflects the contrary” about the Iranian regime.

Sen. John Ossoff, D-Ga., asked Gabbard if there was an “imminent nuclear threat” posed by Iran, referring to stances from the White House and its prior claims about Tehran’s nuclear capabilities being “obliterated.”

In a calibrated answer, she said it’s “not the intelligence community’s responsibility to determine what is and is not an imminent threat” and that the president has authority to make such conclusions.

The Constitution gives Congress — not the president — the authority to declare war, while the president, as commander in chief, directs military operations. But intelligence community analysts and officers frequently compile assessments from a range of sources and methods to inform policymakers, the president and others about the severity of threats.

“You’re evading a question because to provide a candid response to the committee would contradict a statement from the White House,” Ossoff said.

]]>

Golden Dome’s projected cost just jumped $10 billion. Experts fear that’s just for starters.

Thomas Novelly — Tue, 17 Mar 2026 16:20:00 -0400

Golden Dome’s official projected price tag just jumped $10 billion to $185 billion. Experts say the real cost is likely to be far, far more.

Gen. Michael Guetlein, the Space Force officer in charge of the sprawling missile defense project, said Tuesday that it was no longer expected to cost $175 billion, the number given by President Trump when he announced the project last May.

“We were asked to procure some additional space capabilities,” Guetlein told attendees at the McAleese Defense Programs Conference. “So, we are at $185 billion for the objective architecture, which delivers way out into the 2035 timeframe.”

Experts told Defense One it’s most likely going to exceed that figure. Soon after the project was announced, the Congressional Budget Office predicted anywhere from $542 billion to $831 billion over 20 years. In September, an American Enterprise Institute report said the highest-end architecture could mount to $3.6 trillion.

But Guetlein said he was confident in the new figure.

“There's been numerous cost estimates out there in excess of a trillion dollars. I would say the difference between what they are estimating and what we are building is they're not estimating what I'm building,” he said. “We are changing that equation for Golden Dome, simplifying it, if you will, just aggregating it, if you will, to bring down that cost equation and not exceed that $185 billion that the President has committed.”

Todd Harrison, a senior fellow at AEI who has researched Golden Dome’s likely costs, said Guetlein’s $185 billion figure would likely only include a basic capability for space-based interceptors—which has been seen as the most expensive and ambitious part of the project. It’s also, he said, just the beginning of the expenses.

“My understanding is that is just the near-term cost to acquire an initial level of capability. It's not the total acquisition cost, and it does not include long-term operation and replenishment costs,” Harrison said. “It also is a good indicator that space-based interceptors will be a relatively minor part of the architecture, if they even move out of the demonstration phase.”

Last year, the Space Force awarded small contracts to companies to develop orbital interceptors and were soliciting proposals for space-based midcourse interceptors, too. Physicists have questioned the project’s use of boost-phase and mid-course space interceptors, claiming it’s impractical against modern missile threats.

Given Guetlein’s budget estimate, it’s not clear how much space-based interceptors will play a role in the final architecture, said Victoria Samson, the Secure World Foundation’s chief director of space security and stability.

“Really and truly, it's not going to be cheap, but the long pole in the tent in terms of cost is going to be the space-based interceptor layer,” Samson said. “That's going to be the one that's going to drive it up, depending how much they decide to go ahead with it.”

Guetlein admitted that the space-based interceptors are a challenge for the program.

“We have not had anything fail yet, we’ve had numerous successful tests that I can only go into in this environment,” Guetlein said. “If I was to predict where the biggest amount of risk is the space-based interceptor. And it's not the technology, it's the scalability and the affordability.”

The Pentagon is casting a wide net when it comes to Golden Dome-related work. Early this year, the Missile Defense Agency made several announcements that a total of 2,440 applicants have been approved to compete for work totaling up to $151 billion out of an original pool of 2,463, leaving just 23 applicants out of the running.

Guetlein said that Lockheed Martin, RTX, and Northrop Grumman are three prime contractors working with six other companies to build command and control capabilities, or C2, layer for Golden Dome.

“Now, we have a team of nine building our command and control capability, which is really a glue layer that sits upon all these other services and agencies, C2 systems,” Guetlein said.

While Trump’s January 2025 executive order establishing Golden Dome does not establish a due date for the project, a subsequent December 2025 executive order calls for “developing and demonstrating prototype next-generation missile defense technologies by 2028” leaving Guetlein a little more than two years to meet the president’s goals.

“I do not have a 2028 mandate inside the executive order. However, the President did ask us to rapidly change the defensive equation of the nation as fast as we possibly can, and put a marker on there for the summer of 2028,” Guetlein said. “By the summer of 2028 I have to demonstrate the ability with operational capability deployed in the field to defend ourselves against those threats as identified in the executive order. That's what they have asked us to do.”

]]>

Army, Anduril enter into new $20B enterprise agreement

Ross Wilkers — Mon, 16 Mar 2026 17:00:00 -0400

The Army is bringing all 120 of its contracts with Anduril together into a single enterprise agreement with a potential $20 billion ceiling value over up to 10 years.

This pact announced Friday focuses primarily on Anduril’s Lattice operating system, which the company designed as an open software platform that moves data collected from distributed sensors and other feeds into a single integration layer.

"The modern battlefield is increasingly defined by software. To maintain our advantage, we must be able to acquire and deploy software capabilities with speed and efficiency," Gabe Chiulli, the Army’s chief technology officer, said in a release. "Enterprise contracts are a key part of our modernization strategy, allowing us to consolidate software agreements, eliminate redundancies, and accelerate the delivery of critical tools."

For the Army, this agreement is very much in step with the pact it signed with Palantir in August to consolidate 75 contracts into one mechanism. The Palantir agreement has a potential $10 billion ceiling value over up to 10 years.

Both agreements with Anduril and Palantir have an initial five-year base period and a single five-year option. They were set up to establish pre-negotiated terms and pricing, take advantage of volume discounts, and allow the Army to buy goods and services as it needs them.

The Anduril pact covers access to software platforms, integrated hardware, data and compute infrastructure, and a full range of ancillary support services.

Army leaders also prioritized existing data integration with hundreds of systems in the creation of this agreement, which also eliminates pass-through charges on subcontracts.

In entering the Anduril and Palantir agreements, the Army is taking a similar approach to that of the General Services Administration in the OneGov strategy for working more directly with technology providers.

]]>

Defense tech enters a new era: the case of Anthropic and the DOD

Vincent Carchidi — Wed, 04 Mar 2026 12:49:00 -0500

Defense technology often attracts people who are inclined to pay attention to the various stages of a normal technology’s lifecycle: its technical underpinnings and development; its relationship with existing techniques; the intended applications its development seeks to serve; and the process of testing and validating a system’s performance such that it is suitable for real-world deployment and operation. Artificial Intelligence (AI) is no exception.

Recent events indicate that this lens, by itself, is becoming inadequate for understanding the developmental trajectories and practical impacts of AI in U.S. defense. (Editor’s note: Forecast International, like Nextgov/FCW, is owned by GovExec.)

The dispute between Large Language Model (LLM) vendor Anthropic and the U.S. Department of Defense (DoD) is unprecedented. The capabilities of such models are becoming secondary to a broader trend: the relationship between the DOD and private-sector firms working on the frontier of this dual-use technology is characterized by the former’s perception that the latter’s technology is indispensable for their own ends.

This piece provides a thorough timeline of events, including comments on the reported use of Anthropic's "Claude" model in U.S. kinetic attacks in Iran. At the same time, the new position that the DOD has taken on critical dual-use technology like AI is illustrated in connection with the reliability traditionally expected of such technology, arguing that maximum operational access to this technology increasingly takes precedence over these traditional concerns.

What we know so far

Infographic of the Anthropic-DoD dispute timeline by Jorge Morejon

On February 26th, Dario Amodei, CEO of Anthropic, announced the company’s decision in response to the DoD’s ultimatum on the use of the company’s technology: “we cannot in good conscience accede to their request.”

The announcement followed a quick succession of reports indicating that Anthropic and the DOD were at odds over the use of the technology that Anthropic was contracted by the DOD to develop in July 2025.

The following day, in late-afternoon, U.S. President Donald Trump posted on Truth Social, indicating that Anthropic would escape the most significant consequences while nevertheless severing the company’s relationship with the U.S. government: “Therefore, I am directing EVERY Federal Agency in the United States Government to IMMEDIATELY CEASE all use of Anthropic’s technology. We…will not do business with them again!”

The minimum implication of President Trump’s directive — should his post translate into a binding federal directive — is that Anthropic’s contracts with the DOD and broader U.S. federal government were effectively terminated. Likewise, any uses of the models by federal government employees should have ceased.

However, a couple of hours later on the same day, U.S. Secretary of Defense Pete Hegseth posted on X: “…I am directing the Department of War to designate Anthropic a Supply-Chain Risk to National Security…Anthropic will continue to provide the Department of War its services for a period of no more than six months to allow for a seamless transition to a better and more patriotic service.”

Although Secretary Hegseth did not invoke the Defense Production Act, it appears that only one of his directives is possible — either compel other DOD contractors to excise Anthropic’s products from their DoD-related work or continue the use of Anthropic’s models for some period of time.

Amodei responded that night noting that Anthropic had “not yet received direct communication” from the DOD on these decisions, and forcefully stated that they would “challenge any supply chain risk designation in court.”

It is this impending court challenge that will determine the exact course of action taken. Technically, should Secretary Hegseth’s supply chain designation hold before the courts make their final decision, then all DOD uses of Anthropic’s products will begin the months-long process of disentangling these models from DOD workflows and systems. Also in this scenario, DOD contractors would be compelled to excise Anthropic’s products from their work for the DoD, too, though only their DOD work (rather than a given contractor’s work wholesale).

This scenario represents a costly one for Anthropic, as even this temporary designation would disrupt the relationships Anthropic has built with a formidable customer base, particularly in recent months following the release of its Opus 4.5 and Opus 4.6 models.

If Secretary Hegseth’s supply chain risk designation order is stayed by a federal judge as proceedings play out, then the immediate disruption to both DOD and contractor workflows will be delayed. For how long depends on the outcome of the court process. In this scenario, one might expect to see additional jabs at Anthropic made directly or indirectly by the DOD as proceedings unfold. Alternatively, though less likely, an accumulation of pressure from other industry actors might see efforts to reduce the temperature.

The supply chain risk designation and six-month grace period are, nevertheless, seemingly incompatible; it is unclear how a federal court will navigate them.

OpenAI’s DOD maneuver

As this saga was unfolding, an open letter was published last week by employees of OpenAI and Google — but not formally associated with either company — expressing support for the “red lines” that Anthropic established in its relationship with the DoD, and urged their companies’ leaders to “refuse the Department of War’s current demands for permission to use our models for domestic mass surveillance and autonomously killing people without human oversight.”

OpenAI CEO Sam Altman internally responded to his staff’s concerns on the 27th, before President Trump’s post, noting his intention to both retain the guardrails Anthropic sought on its own models when used in classified settings and to “de-escalate” the Anthropic-DoD dispute.

OpenAI is, for what it is worth, substantially less motivated by safety concerns than rival Anthropic (a difference rooted in its history). Altman specifically is an (in)famously skilled operator, having regained his position as CEO of OpenAI after a brief termination by the OpenAI Board of Directors in 2023.

Perhaps unsurprisingly, then, Altman posted on X on the night of the 27th that OpenAI closed a deal with the DOD to make OpenAI’s models available on classified networks. Bizarrely, Altman included the following detail: that the DOD accepted OpenAI’s “red lines,” which Altman suggested other AI firms — like Anthropic — would accept.

Amodei previously confirmed that Anthropic’s red lines were: (1) The use of its AI products for mass domestic surveillance (which is explicitly distinguished from “lawful foreign intelligence and counterintelligence missions”); and (2) The use of its AI products for fully autonomous weapons (which is explicitly distinguished from “partially” autonomous weapons currently in use in Ukraine).

It is unclear why the DOD would accept these red lines in the case of OpenAI but not Anthropic, save for some differing interpretation therein.

Some minor clarity is provided by OpenAI’s February 28th statement on its DOD deal, where it lays out three red lines to which the DOD is said to have agreed (below is a direct quote):

No use of OpenAI technology for mass domestic surveillance.
No use of OpenAI technology to direct autonomous weapons systems.
No use of OpenAI technology for high-stakes automated decisions (e.g. systems such as “social credit”).

OpenAI distinguishes these red lines from reliance on a company’s usage policies to guide defense deployments, noting that the agreement specifies deployment of OpenAI’s models is exclusively cloud-based and accompanied by the company’s “full discretion over our safety stack…” An excerpt of the contract is provided in which DOD Directive 3000.09 on autonomy in weapon systems is referenced as a standard for the usage of OpenAI’s models.

Notably included in this statement is a blunt, single-sentence opposition to the designation of Anthropic as a supply chain risk.

Interestingly, Anthropic made a change to its Responsible Scaling Policy that voluntarily governs the company’s efforts to mitigate catastrophic AI risks on February 25th. The new policy does away with the prior (self-imposed) restrictions on the training of AI models whose risks cannot be fully mitigated in advance, indicating that the company likely believed there was room for compromise with the DOD before OpenAI’s maneuver.

This shift is consistent with reporting that Anthropic was negotiating with the DOD through last week, understanding themselves to be on track for a deal, right until the moment of President Trump's announcement on Friday. (Anthropic had privately rejected a proposal to keep their models on the cloud rather than instantiated in the weapon itself.)

Altman has since suggested OpenAI's deal was "rushed." OpenAI lead researcher Noam Brown posted on X on March 3rd:

@OpenAI will not be deploying to the NSA or other DoW intelligence agencies for now, so that there's time to address potential surveillance loopholes through the democratic process.

Note that Brown's comments target intelligence agencies, and not lethal autonomous weapon systems (LAWS).

Importantly, there is no indication that OpenAI's contract is terminated, nor is it clear that a substantive adjustment to its language is immediately forthcoming based on this announcement, beyond certain language already slated for an amendment. The timeline for intelligence gathering or analysis using OpenAI's models is therefore an open question.

In any event, OpenAI’s maneuver into the DOD in part serves to further embed OpenAI’s models in classified networks where Anthropic previously held a firm lead.

Reliability: second to operational access?

The past week or so has seen discussions of LAWS with reference to LLMs. It is often unclear what commentators mean by this, as such ideas are, largely, hypothetical.

Still, the past week has shown that traditional DOD concerns with the reliability of mission- and safety-critical defense technology - like LAWS - have been secondary to operational access therein.

Among the more striking aspects of Amodei’s February 26th announcement is a frank admission: the technology the company develops — long-term visions notwithstanding — is not reliable enough to safely operate in domains where fundamental civil liberties or life itself are at stake. To be sure, the qualifier “today” accompanies these remarks. But consider Amodei’s remark:

But today, frontier AI systems are simply not reliable enough to power fully autonomous weapons…fully autonomous weapons cannot be relied upon to exercise the critical judgment that our highly trained, professional troops exhibit every day. They need to be deployed with proper guardrails, which don’t exist today.

This comment is all the more striking when one realizes Amodei need not have made it — the company is riding a commercial high with a product - Claude Code - of some significant popularity within the DOD and elsewhere. Amodei could have opposed the use of Anthropic’s technology for LAWS on moral grounds without including this particular detail.

Amodei’s comment was echoed by U.S. Air Force Gen. Jack Shanahan (ret.), who wrote on LinkedIn:

No LLM, anywhere, in its current form, should be considered for use in a fully lethal autonomous weapon system…Despite the hype, frontier models are not ready for prime time in national security settings. Over-reliance on them at this stage is a recipe for catastrophe.

These views are relatively in sync with one another, revolving around the suitability of a given class of AI models (transformer-based LLMs) for a given set of tasks (those that are safety- or mission-critical in nature).

LLMs do not provide guarantees on their performance. This is a banal fact about such models. This fact is routinely band-aided through scaffolding in which the LLM is situated in a structure that may have a separate and domain-specific (typically limited) model verify its outputs; prompting techniques to guide the LLM to more appropriate outputs; or wholesale restrictions on their use in tasks where the margin for acceptable error is essentially zero.

The general point has reverberated throughout the DOD and industry. It is why, in August 2025, Palantir Chief Revenue Officer Ryan Taylor bluntly stated: “LLMs, on their own, are at best a jagged intelligence divorced from even basic understanding.” In February of this year, U.S. Air Force Lt. Col. Matthew Jensen suggested that an LLM could conceivably be used as an interface with which human pilots can debrief on the activity of a Collaborative Combat Aircraft’s autonomy suite, implying a separation of critical systems.

Interestingly, although the DOD is traditionally concerned with the reliability of technologies deployed in mission-critical domains, it is primarily Anthropic, and not the DoD, that has expressed this concern in the current context. Rather than indicating that the DOD is unconcerned with such matters, it is more likely that the locus of dispute as perceived by DOD leadership is not the question of whether Anthropic’s technology is sufficiently reliable, but the question of whether the DOD has maximum operational access to it.

Operational access to Claude and the Iran strikes

Consistent with this view, The Wall Street Journal reported on Sunday that the DOD had used Anthropic’s Claude model in support of the U.S.’s attacks on Iran (undertaken in coordination with Israel). The report notes that U.S. Commands globally, including U.S. Central Command - with the Middle East under its purview - use this tool. For Iran specifically, it states that Central Command used Claude for “intelligence assessments, target identification and simulating battle scenarios…”

To what extent Claude informed these activities and substantively shaped their consequences is unclear, as was the case for similar reporting on Claude’s use in the capture of former Venezuelan President Nicolás Maduro (e.g., there is a range of possible ways in which such a tool could be used for intelligence assessments, varying dramatically in their practical significance).

Note, too, that this report indicates that Anthropic's supply chain risk designation was not in force at the time of these strikes, or it was circumvented - a potential fact of relevance should Anthropic proceed with their lawsuit. As of March 2nd, U.S. cabinet agencies began ceasing their uses of Anthropic's AI products in favor of OpenAI's and Google's. The State Department reportedly cited President Trump's directive as justification.

That said, it is fascinating that Anthropic’s model would be used in target identification. Coupled with the lack of performance guarantees offered by LLMs, and the relatively limited information available, this use bears some resemblance to the Israeli Defense Forces' (IDF) own use of AI in target acquisition.

The IDF expressed concern over the past decade (at least) for the ability to acquire targets efficiently during wartime, born of the Israeli Air Force’s “shortage” of identifiable and valuable targets in prior conflicts. The General Staff Targeting Directorate was established in 2019 explicitly to link data science and machine learning for this end.

The use of two AI systems (of some largely unspecified but likely simple designs) called “Hasbora” and “Lavender” were recruited by the IDF after the October 2023 Hamas attack for the acquisition of infrastructure targets and individual targets in Gaza, respectively.

Just to say: the use of AI in target acquisition and other activities during extended combat operations is not separable from goals and standards for execution therein. Whether Anthropic’s tools were significantly used in this context is only in part a question of the model’s capabilities; the other side of this coin is whether these human ends require the reliability Amodei suggested is unavailable, a determination exclusively made by the human personnel and commanders involved. The technology does not itself set these standards, and could not enforce those that exist.

That said, sharing information that Claude was used directly in the service of kinetic attacks — although not breaching the original red lines set out by the company — might be understood as a kind of psychological warfare against Anthropic itself, as the DOD asserts its access to these models. Researchers at this firm routinely characterize their work developing Claude as one might when cultivating virtues in an impressionable human. A psychological effect on them would be anticipated by those within the DOD when sharing the news with reporters.

Defense tech's future beckons

Secretary Hegseth’s apparent willingness to designate Anthropic, an American defense contractor, a supply chain risk is unprecedented. Whatever the outcome of any impending court challenge, the DOD has effectively indicated a willingness to go to extreme lengths in gaining access — both formal and operational — to state-of-the-art, dual-use defense technology. The stakes and implications of this development are significantly greater than earlier precedents for DoD-Silicon Valley tumult, including Project Maven.

This dispute is likely to continue reverberating throughout the defense technology and policy worlds. Indeed, one finds Dean Ball, a lead drafter of the Trump administration's AI Action Plan, writing in reference to this dispute with dread.

A Frankenstein’s Monster effect, in which the developers of LLMs have found their pronouncements on the world-historical nature of their technologies acceded to, now defines their defense relationships in ways those same developers likely did not anticipate.

]]>

Strikes on Iran will test US cyber strategy abroad, and defenses at home

David DiMolfetta — Sat, 28 Feb 2026 18:16:26 -0500

Coordinated U.S. and Israeli strikes on Iranian targets are putting renewed focus on how the United States integrates offensive cyber capabilities into the battlespace — and how prepared federal agencies are for retaliation at home.

Iran has shown a tendency to respond to overseas threats with cyber means, from defacing websites to spying on U.S. and allied targets. Tracking such actions and alerting the U.S. government and public is a job of the Cybersecurity and Infrastructure Security Agency, which has been operating with sharply reduced staffing due to a funding lapse for its parent agency, the Department of Homeland Security.

“This is a bad time for Washington’s cyber agency to be operating with limited staff,” said Annie Fixler, director of the Center on Cyber and Technology Innovation at the Foundation for Defense of Democracies, a national security think tank.

That funding lapse comes after Trump-administration moves shrank CISA’s workforce by about one-third last year and degraded public-private collaboration mechanisms. This “limits the ability of the federal government to provide timely cyber threat information to the private sector,” Fixler said.

In the wake of the U.S. and Israeli airstrikes, American companies could see a “barrage” of low-level attacks like website defacements and distributed denial-of-service attacks, said Fixler. “Iran might also see some limited success against targets that do not have proper cyber hygiene — exposed edge devices with default passwords, for example.”

Other cyber experts said the U.S. should prepare for a mix of distributed denial-of-service campaigns, ransomware and hack-and-leak operations meant to send a message.

“While it’s not operating at the same technical level as China or Russia, Iranian-linked groups have carried out disruptive attacks against U.S. financial institutions, infrastructure providers and private sector companies,” said Tom Pace, a former Marine intelligence specialist and CEO of NetRise, a cybersecurity supply chain firm.

The conflict will likely see a surge in state-sponsored hacking activity, “specifically targeting operational technology and critical infrastructure through the exploitation of internet-facing industrial control systems and vulnerable [programmable logic controller] hardware,” said Brian Harrell, a former CISA official.

“Threat hunters should be working overtime right now. By combining disruptive attacks with psychological operations, Iran will seek to erode public trust in government institutions and project domestic strength during periods of heightened conflict,” he said.

Elisity CEO James Winebrenner echoed that advice. “We should be vigilant in protecting exposed [industrial controls systems] and expect heightened retaliatory activity in the coming days and weeks,” he said. In late 2023, Iran-linked hackers digitally defaced U.S. water treatment equipment.

Tehran may play up the effectiveness and scope of their cyberattacks, said Cynthia Kaiser, a former FBI cybersecurity deputy director who leads the Ransomware Research Center at Halcyon. Industry research has documented these theatrics.

“They’ll turn [an intrusion] into an information operation, and say, ‘Look, we compromised this entire facility,’ even though they compromised just a machine,” Kaiser said.

Asked about the diminished DHS and CISA workforce, Kaiser said other national security elements across the government like the FBI and NSA are still able to track and respond to cyber threats in full. “People marshal themselves together to focus on a big threat” even if there are resource shortages, she said.

Matt Hayden, a former DHS infrastructure security official, said CISA would continue its standard threat-hunting procedures as if the government was fully operating. “While there are operators that are working without pay, they are still working,” he said. Hayden is now vice president of cyber and emerging threats at GDIT.

Defense One has asked CISA and DHS for comment.

The U.S. has likely deployed a powerful toolset of cyber and electronic operations against Iranian targets, said Charles Moore, a retired three-star general and former U.S. Cyber Command official who is now a distinguished visiting professor at Vanderbilt University’s Institute of National Security.

“I would suspect that anything that Iran is using to communicate, anything they’re using to keep situational awareness or visibility on the battle space, and any systems they’re using to try to defend themselves, all those types of things — would be targets that would be of interest from a cyber perspective,” Moore said.

The U.S. and Israel are also likely intercepting communications to aid in its operations. “In general, signals intelligence of any type, is something the United States is very interested in and is very adept at gathering. And so I have no doubt that those types of efforts will continue,” he said.

Internet connectivity in Iran has also been heavily reduced. The exact cause of this decline is uncertain. While the U.S. or Israel may have played a role, Iran frequently restricts internet access during periods of unrest, such as anti-regime protests.

In coming days, there may be public indications that Cyber Command played a role in U.S. components of the operation, said FDD’s Fixler.

Influence operations have played a role in the efforts. Israel notably hacked a major Iranian prayer app, aiming to fuel uprising against the regime. But its effectiveness may be limited, said Maggie Feldman-Piltch, CEO of Iceberg Holdings, a firm that helps private-sector entities prevent IP theft.

The infiltration of a prayer app with those messages is “a wonderful example of not knowing your audience or understanding what happens when you don’t,” said Feldman-Piltch, who formerly led the digital and electronic portfolio at the Wilson Center.

A simple message finally calling for uprising ignores years of already documented protests against Iran that have resulted in civilian killings, she said.

The U.S. and its allies will have to stay vigilant. The operation “has destroyed Iran’s conventional military options, making cyber operations the regime’s sole remaining instrument of asymmetric retaliation,” says a threat intelligence report sent to Defense One produced by cybersecurity firm Anomali. Iran-linked cyber units were “activated and retooling before the kinetic trigger,” it adds.

“Geography provides no protection against a cyber-enabled adversary,” said Tatyana Bolton, principal and head of Monument Advocacy’s cybersecurity practice. “Iran possesses some of the most creative and dangerous cyber operators in the world, and with the current escalation, their incentive for restraint is significantly reduced.”

“They don’t need to win a naval battle in the Gulf to hurt the U.S. — they can simply hold our power grids, water systems, and hospitals hostage from halfway around the world to force our hand at the negotiating table,” Bolton said. “We must recognize that in 2026, the front line isn’t just in the Middle East — it’s in our own backyard.”

]]>

Trump directs government to ‘immediately cease’ using Anthropic technology

Frank Konkel — Fri, 27 Feb 2026 18:26:00 -0500

President Trump on Friday directed all federal agencies — including the Defense Department — to “immediately cease all use” of frontier AI firm Anthropic’s technology.

Trump’s announcement followed a tense back-and-forth between Anthropic and the Pentagon, which widely uses the San Francisco company’s popular AI platform, Claude, in even classified networks but took issue with the company’s refusal to give the Pentagon unrestricted access to its models.

In a Thursday statement ahead of the Pentagon’s Friday deadline, Anthropic CEO Dario Amodei said he refused to allow Claude to be used for mass surveillance of U.S. citizens or to guide fully autonomous weapons, an argument Trump framed as trying to “strong arm” the Defense Department and force it to “obey their terms of service.”

“I am directing every agency in the United States Government to IMMEDIATELY CEASE all use of Anthropic’s technology,” Trump said in a Truth Social post. “We don’t need it, we don’t want it, and will not do business with them again.”

Trump said there would be a six-month “phase-out period” for agencies using Anthropic’s products at various levels, including classified settings and among civilian agencies. Trump threatened Anthropic with punishment should the company refuse to help in the phase-out. As Defense One’s Patrick Tucker reported Feb. 26, it may take several months or longer for the government to replace Anthropic’s tools.

“Anthropic had better get their act together and be helpful during this phase out period, or I will use the full power of my Presidency to make them comply, with major civil and criminal consequences to follow,” he said.

In his own post, Defense Secretary Pete Hegseth said he was ordering his department to “designate Anthropic a Supply-Chain Risk to National Security.”

Hegseth did not explain why a supply-chain risk would be permitted to operate in classified networks for six more months.

Amodei had noted this “contradictory” action in his Thursday statement.

“They have threatened to remove us from their systems if we maintain these safeguards; they have also threatened to designate us a ‘supply chain risk’ — a label reserved for US adversaries, never before applied to an American company — and to invoke the Defense Production Act to force the safeguards’ removal. These latter two threats are inherently contradictory: one labels us a security risk; the other labels Claude as essential to national security.”

Founded in 2021, Anthropic has developed models and tools that are already widely used across the federal government, largely through its partnership with leading cloud provider Amazon Web Services, through which it first gained a foothold in the Defense Department and intelligence agencies. Anthropic, along with xAI, Google and OpenAI, received $200 million defense contracts last July to bolster the Pentagon’s push to harness AI.

Meanwhile, the General Services Administration, which manages hundreds of billions of dollars’ worth of contracts on behalf of all agencies, said in a statement Friday it would remove Anthropic from its Multiple Award Schedule and USAI.gov. Federal Acquisition Services Commissioner Josh Gruenbaum tweeted that GSA has terminated Anthropic's OneGov deal, ending the availability of those contracts across the Executive, Legislative and Judicial branches.

“GSA stands with the President in rejecting attempts to politicize work dedicated to America’s national security,” GSA Administrator Edward C. Forst said in a statement. “Building resilient, secure, and scalable AI solutions demands alignment, trust, and a willingness to make hard calls. We’re committed to delivering results for Americans, and working with our AI industry partners who fit the bill.”

The rhetoric used by Trump, Hegseth, Pentagon spokesperson Sean Parnell, and Defense Undersecretary for Research and Engineering Emil Michael was notable for its stridency.

Hegseth posted: “...@AnthropicAI and its CEO @DarioAmodei, have chosen duplicity. Cloaked in the sanctimonious rhetoric of ‘effective altruism,’ they have attempted to strong-arm the United States military into submission - a cowardly act of corporate virtue-signaling that places Silicon Valley ideology above American lives...”

Michael posted, “...It’s a shame that @DarioAmodei is a liar and has a God-complex. He wants nothing more than to try to personally control the US Military and is ok putting our nation’s safety at risk...”

And yesterday, Parnell posted that DOD only seeks the ability to “use Anthropic's model for all lawful purposes,” adding that the idea that the Pentagon wants fully autonomous weapons or mass surveillance is a false narrative “peddled by leftists in the media.”

But in his statement, Amodei said those are the only two limits he insists on.

In “a narrow set of cases, we believe AI can undermine, rather than defend, democratic values. Some uses are also simply outside the bounds of what today’s technology can safely and reliably do,” he said in his statement.

In a statement, Anthropic said it has "not yet received direct communication from the Department of War or the White House on the status of our negotiations."

"We have tried in good faith to reach an agreement with the Department of War, making clear that we support all lawful uses of AI for national security aside from the two narrow exceptions above," the company said. "To the best of our knowledge, these exceptions have not affected a single government mission to date."

Editor's note: This story was updated to include a statement from Anthropic.

Bradley Peniston contributed to this report.

]]>

Now you can train for the next drone war on simulated Ukrainian front lines

John Breeden II — Tue, 23 Dec 2025 08:00:00 -0500

The war in Ukraine has made one thing unmistakably clear: drones are no longer a support weapon on the battlefield. They are central to how modern wars will be fought. From reconnaissance and targeting to direct strikes, low-cost, fast-moving drones with powerful payloads are causing massive destruction and racking up kills against traditional armed forces. And that mastery of drone warfare is allowing a smaller country like Ukraine to hold its own on the front lines against a superpower, performing devastating strikes that keep the Russian Army on its back foot.

The world is taking notice. Other countries in the region are already planning to use drones in their defense should war come to their door, while also countering those of their potential enemies. The shift in military thinking was on display earlier this year as Estonia announced plans to build a so-called “drone wall” along its border, a layered defensive system designed to detect and counter unmanned aerial threats. As Estonian defense officials explained to Nextgov/FCW, their goal is not simply to buy more hardware, but to rethink how borders and battlespaces are protected in an era where inexpensive drones can have a large strategic impact.

But that same lesson is not theoretical in Ukraine, where drones have become one of the most decisive tools of the conflict. In Ukraine, small, fast and often expendable drones are now used for everything including reconnaissance, precision attacks in the midst of combat, electronic warfare, air defense interception and even strategic strikes far behind the frontlines. Most of the drones used by Ukraine are FPV, or first person view, models flown by operators who often receive only weeks of training before being sent to the front. In that environment, pilot skill matters a lot. And that is where simulation technology has quietly become just as important as the hardware.

Over the past two years, Ukrainian forces have relied heavily on high-fidelity drone simulators to train operators before they ever touch a real aircraft. According to Simtech Solutions, the developers of one of the main Ukrainian training tools, their platform alone has been used to get over 5,000 military pilots flight ready, capable and certified to perform even the most advanced and critical missions. And the results speak for themselves. The company says that pilots who trained with their simulator have successfully struck over 100,000 real-world targets.

That platform has now been released in a public-facing version called Ukrainian Fight Drone Simulator, which is available on the Steam platform. I was able to experiment with the new simulator, logging in over 100 hours of flight time and training. With real-world physics and battlefield conditions, it was not a walk in the park. I ended up crashing almost 200 drones before I even got through the training academy, much less get anywhere close to an enemy. But even while I was still learning how to fly drones in combat, it was clear how much this technology could dominate the battlefields of the future, perhaps even more so than any other conventional weapon.

Although it’s available on a gaming platform like Steam, calling Ukrainian Fight Drone Simulator a game understates its rich history on the frontlines and its commitment to realism. The developers describe it as a public adaptation of their ultra-realistic trainer, built directly from battlefield lessons. Everything from the physics of flight to the ruthlessness of battlefield opponents are intentionally unforgiving.

You can’t even use a game controller too effectively with it. For a true training experience, you need to have a real RC Controller, just like the pilots in Ukraine. I used my Radiomaster TX16S that controls my personal drone, but almost any other model works just fine. The developer recommends Radiomaster Pocket, Zorro, TX12, Boxer, TX16S or TBS Tango 2 RC controllers.

The simulator includes multiple classes of combat drones, as most of the ones being used in Ukraine are fairly modular in terms of their possible loadouts. You can set them up for bomb drops against infantry, although that technique can also be effective against light vehicles or rear supply areas. More powerful are the kamikaze drones that can lug around huge bombs or artillery shells in their undercarriages. Although this greatly complicates flight physics, direct hits from them can destroy armored vehicles — like fast-moving Russian BMPs — or hardened targets — like earthen bunkers. Even frontline tanks, once the kings of the battlefield, can be damaged or disabled by a hit from a drone in a critical spot, and then destroyed with follow-up strikes.

If you are up for a real challenge, air interceptor drones are also included. If you thought that hitting a moving vehicle or a group of running soldiers was difficult, try chasing down a helicopter or even an enemy drone. It takes a lot of skill and practice.

What really stood out to me during my time on the simulated frontline was how much Ukrainian Fight Drone Simulator emphasizes military elements beyond just flying drones. You actually have to plan your missions, taking into account things like time of day, lines of sight, potential exposure to enemy defenses and whether or not the enemy is operating electronic warfare in the area. I had to learn how to manage things like battery power too, because going around an area with dense anti-air defenses to get at juicy targets sitting behind them might be a great idea, but only if your heavily laden drone has enough juice to make the run and still give you a least a few minutes of time over your target.

Although I had a great experience with Ukrainian Fight Drone Simulator, it was not really about flying for fun. It was more about understanding how drones can be successfully used in modern conflicts. My success in the field really depended on training, discipline and repetition.

And that is where something like Ukrainian Fight Drone Simulator becomes relevant far beyond Ukraine. For defense agencies, allied militaries and even policymakers, tools like this provide insight into why drones are changing warfare so rapidly. They also help explain why training has become such a critical bottleneck. Hardware can be produced quickly. Skilled operators cannot.

What simulations like Ukrainian Fight Drone Simulator ultimately demonstrate is that the drone revolution is not theoretical. It’s operational, and it’s happening now. Armies that go into their next fight without an extremely capable drone wing and a cadre of qualified and experienced pilots are likely to be at a serious disadvantage.

As more agencies track drone developments and consider how to defend against or deploy them, simulations will play an increasingly important role too. They offer a safe way to understand a technology that is anything but safe on the battlefield. And for anyone trying to grasp how modern conflicts are evolving, stepping into a virtual cockpit like the ones that Ukrainian Fight Drone Simulator offers may be the most instructive place to start.

John Breeden II is an award-winning journalist and reviewer with over 20 years of experience covering technology. He is the CEO of the Tech Writers Bureau, a group that creates technological thought leadership content for organizations of all sizes. Twitter: @LabGuys

]]>

More AI tools coming in days or weeks, Pentagon R&D chief says

Patrick Tucker — Tue, 09 Dec 2025 10:33:00 -0500

Updated Dec. 9, 10 a.m. ET:

The Pentagon will widely deploy new AI tools for logistics, intelligence analysis, and combat planning in days or weeks, its research-and-engineering chief said Monday, adding that wide deployment of artificial intelligence now tops his list of “critical technologies.”

The department has chosen Gemini for Government as the platform that will support DOD’s first department-wide rollout of AI tools, Google and defense officials announced Tuesday morning.

The moves come after the Defense Innovation Unit, the Chief Digital and Artificial Intelligence Office, or CDAO, and others were combined under Emil Michael, defense undersecretary for research and engineering, in a bid to accelerate deployment of AI and other technologies. He said that he will likely reduce the number of technology areas that DIU is working on as well.

The advent of large-language-model tools such as ChatGPT, Claude, and Gemini have made it possible—and necessary—to develop AI tools faster, Michael told reporters at the Defense Writers Group on Monday.

“The explosion of capabilities has been enormous, and we're just catching up to that,” he said. “Now we can take CDAO and actually try to use it to push the capability into the Department for actual use cases.”

He said that expands the usefulness of CDAO, which was largely managing in-house analytic tools like Advana and exploring data assets within the military.

The explosion of ChatGPT and other consumer tools makes that necessary, Michael said Saturday during the Reagan Defense Forum in California.

“For a department of 3 million people, we're vastly under-utilizing AI relative to the general population,” he said.

On Monday, Michael said Russia’s war on Ukraine and Ukraine’s response serves as a key lens on future conflict.

“You have a robot on robot frontline now, which we've never seen before,” he said.

And China’s military buildup of the past 10 to 15 years—“the most significant” in world history, he said— also “requires a kind of a different mindset.”

Michael said China is working to reverse-engineer advanced chips and to develop its own.

“China is absolutely trying to indigenize their own TSMC. If you look at the supply chain of the ASML, TSMC, and Nvidia, [China is] trying to replicate that capability with their own domestic sources,” he said.

Michael said he is seeking help from foreign countries such as Australia and South Korea, searching for more sources for chips, access to test ranges for hypersonic weapons, and more.

Last month, he pared the list of critical technology areas that his office would pursue from 14 to six. (“14 priorities, in truth, means no priorities at all,” he said in a Nov. 17 video.) On Monday, he said that wide AI deployment would be his top priority.

“I'm going to put the capability in front of you so you can start learning, using it. We'll have training. We'll have support for deployed engineers, all that. And then you'll see innovations come from there.”

He said his office would soon announce acquisition changes along the lines of the Pentagon’s broader November announcement.

Michael made his remarks days after the White House released a National Security Strategy that declared an intention to refocus the U.S. security and strategy toward the Western Hemisphere. He deferred policy questions to people “abiove his pay-grade’ but said that he was still “focused much more on other parts of the world”—particularly China, the potential adversary whose capabilities are closest to the Pentagon’s.

]]>

Hegseth broke rules, DODIG concludes, even though he said Yemen strike details were ‘safe to declassify’

Meghann Myers — Fri, 05 Dec 2025 12:00:00 -0500

Defense Secretary Pete Hegseth told the inspector general investigators looking into his alleged use of Signal to share classified strike plans that he determined the details he shared either weren’t classified to begin with or were “safe to declassify,” according to a written statement included in a report released Thursday by the Defense Department’s independent oversight office.

Though Hegseth had the authority to make that decision, the IG found he was still in violation of department policy for transmitting information that could have put service members in danger—in this case, the Navy pilots who were flying the March 15 mission to bomb Houthi targets in Yemen—by taking sensitive details about time, place and manner and sharing them on an unapproved messaging platform via his personal cell phone.

“The Secretary sent nonpublic DOD information identifying the quantity and strike times of manned U.S. aircraft over hostile territory over an unapproved, unsecure network approximately 2 to 4 hours before the execution of those strikes,” the report found. “Using a personal cell phone to conduct official business and send nonpublic DOD information through Signal risks potential compromise of sensitive DoD information, which could cause harm to DOD personnel and mission objectives.”

The report finds several issues with Hegseth’s communications, namely that while declassified, the strike plans were still very much sensitive information, and that he used an unapproved app on his personal cell phone to communicate them, then did not take care to retain what is clearly official correspondence.

“The United States military has successfully managed operations for two and a half centuries without the use of an instant messaging application, and why the secretary felt the need to use that application—while sitting in the SCIF and with access at his fingertips to the two more appropriate communication tools—he chose to use this unapproved tool is mysterious to me, right?” Greg Williams, the director of the Center for Defense Information at the Project on Government Oversight, told Defense One.

The move calls Hegseth’s judgment into question, as the declassification of information doesn’t necessarily mean that it’s fit to be broadcast in a group chat whose members, it would appear, were not all accounted for as intended recipients.

“It's like buying something inappropriate on your expense account and claiming that because the expense account was approved, or didn't violate any written rules, that the expense was necessarily prudent,” Williams said.

Hegseth’s spokesman on Tuesday posted on X that the report is a “TOTAL exoneration.”

That’s not what the report says.

“Because the Secretary indicated that he used the Signal application on his personal cell phone to send nonpublic DOD information, we concluded that the Secretary’s actions did not comply with DOD Instruction 8170.01, which prohibits using a personal device for official business and using a nonapproved commercially available messaging application to send nonpublic DOD information,” the report reads.

Additionally, DOD officials are required to keep all official correspondence. The IG had to use the Atlantic’s transcript of the Signal chat because the messages in Hegseth’s phone had auto-deleted.

The IG recommends that U.S. Central Command, which sent the classified email from which Hegseth pulled the details, review its classification procedures to ensure it prints classification warnings after each paragraph of a document, instead of just at the top of it (as was the case with the March 15 plans).

A separate IG report recommends further training on the use of personal devices, as well as a department-wide review of their use to conduct official business.

Hegseth’s written statement to the IG, given in lieu of the requested sit-down interview, provides some of the first insight into the decision-making that went into sharing strike plans on the Signal group chat, which included national-security officials, but also fatefully, the editor of The Atlantic.

In public statements, both Hegseth and his spokespeople have insisted that he shared no classified information, but have never included the explanation that Hegseth had thoughtfully declassified the information before he shared it. There is no official process for a defense secretary to declassify information, so Hegseth deeming it so is as legitimate as anything else.

Hegseth vs. IGs

Inspector general reports are not in the business of handing down punishments, so it will be entirely up to President Trump whether Hegseth faces any repercussions.

As of press time, the secretary had not made any public statements in response to the report’s public release. Spokespeople for the secretary did not immediately respond to a request for comment.

But Hegseth has already taken steps to constrain his department’s IGs, though he has stopped short of adding to the Trump administration’s unprecedented firings of IG at nearly two dozen other agencies. During his Quantico speech to flag officers in September, he said the independent investigators had somehow been “weaponized.”

“I call it the ‘No more walking on eggshells policy,’ ” Hegseth told the auditorium full of senior leaders. “We are liberating commanders and NCOs. We are liberating you. We are overhauling an inspector general process, the IG that has been weaponized, putting complainers, ideologues and poor performers in the driver's seat.”

His guidance to the service secretaries included tightening the threshold for opening IG investigations, requiring frequent written updates, and creating a system for tracking “serial complainants”—a problematic step because IG hotlines are anonymous.

“It was not clear to me then, and it is still not clear to me now, whether Hegseth intended for that to apply to the military branch IGs, the DOD IG, or both,” Faith Williams, who directs the Effective and Accountable Government Program at the Project on Government Oversight, told Defense One.

So far, Hegseth has only issued orders to the military departments to overhaul their local IG processes. In the case of his own department IG investigation, it was requested by the heads of the Senate Armed Services Committee.

“But I think regardless of who he ultimately intends those changes to impact, what was really clear reading through that memo was this kind of dripping disdain for whistleblowers and for the work that an inspector general office does,” Williams said. “And I think we can all agree that, let's work together to make improvements to IG processes and procedures and make sure we have the best people serving in those roles. I am interested in that, too. I'm not saying IGs are above reproach, but that memo just really dripped with hostility toward whistleblowers, and it does beg the question, ‘why?’”

]]>

The Army is taking counter-drone experimentation from Europe to INDOPACOM

Meghann Myers — Tue, 25 Nov 2025 16:41:00 -0500

U.S. soldiers deployed to Europe had a busy November testing out counter-drone systems that the service hopes to get into the hands of more NATO allies, as well as with units and allies as far away as the Indo-Pacific.

First Polish, Romanian, and American troops trained together Nov. 18 in Poland on Merops, an AI-enabled, pickup-truck-transportable system that identifies enemy drones, then launches a cheap fixed-wing drone to ram them. At the same time, the Army held Operation Flytrap 4.5 in Germany, a competition of 20 cUAS contenders in a competition for one of four $350,000 prizes.

“It also demonstrated our capability, just as Flytrap did, to integrate with industry, to move very quickly to employ a capability that's lethal,” Brig. Gen. Curt King, who leads the 10th Army Air and Missile Defense Command in Germany, told reporters Tuesday. “It can defeat the Shahed-type threats, but also it demonstrates our ability to place capabilities that are much cheaper than some of our other previous systems that we've been using to date, to ensure that we are able to build the capacity against the drone threats that could be placed into the air.”

The U.S. has been using air defense systems to shoot down drones, with missiles that cost millions of dollars each. A Merops interceptor drone costs about $15,000, about half the price of the Shaheds that Ukrainians have been shooting down with it.

“The other thing that we demonstrated with Flytrap, and that Ukraine has shown us … is the technology is rapidly evolving so that we can get to enhanced decision aids and autonomy, which Ukraine has been rapidly developing, so that I don't need 10 soldiers to do a function,” King said.

Both events included the team from the Global Tactical Edge Acquisition Directorate, a nascent procurement system that plans to create a marketplace where units can buy the systems that are vetted and approved through events like Project Flytrap.

“We're not just stopping with the counter-UAS fight,” Col. Christopher Hill, senior director of the GTEAD, told reporters. “Next up on the list is ground autonomy and ground launched effects, as I mentioned earlier, the offensive systems that we're going to use to create a dilemma for our adversaries. Then we're going to move to air autonomy and air-launched effects—again, another offensive system to provide a dilemma.”

The team is expanding to Indo-Pacific Command early next year, Hill said.

“We're going to replicate the same processes there in the Pacific region, in order to not only support our U.S. unit there, but also our international partners in Australia, in South Korea and in Japan, and other partners there in the region,” he said.

Beyond demonstrations, GTEAD plans to have soldiers lead assessments and give feedback so that systems can be tweaked and then put on the marketplace.

“On the back end of these demonstrations, be prepared, from an acquisition standpoint, to actually put dollars towards these capabilities and give these companies something to look forward to from a contract standpoint,” he said.

]]>

Foreign spies are targeting Army soldiers, civilians and families, official warns

David DiMolfetta — Mon, 24 Nov 2025 15:13:00 -0500

A top military intelligence official issued a memo this month warning Army servicemembers of foreign adversaries’ continued attempts to gather intelligence by targeting personnel, including civilians and their families, via fake companies and recruiters.

The Nov. 13 message authored by Army deputy chief of staff for intelligence Lt. Gen. Anthony R. Hale was sent to over a million soldiers and civilians in the Army apparatus and noted the threat of foreign intelligence entities trying to gather information remains persistent, the Army said in a Monday statement.

The information, which was voluntarily publicized, underscores that foreign rivals are continuing covert online efforts to access national intelligence and defense information from people connected to the U.S. government, including military members. It also comes as the Trump administration has worked to severely shrink the government through layoffs and paid offers for employees to leave federal service early.

Foreign spy groups are posing online as consulting firms, corporate recruiters, think tanks and other seemingly legitimate companies, Hale’s message said. It does not name specific nations involved.

“Especially in the context of the recent lapse in appropriations and government shutdown, our adversaries are looking online to identify individuals seeking new employment opportunities, expressing dissatisfaction or describing financial insecurity,” he wrote. The government reopened on Nov. 12, just before the memo was issued.

Current and former federal workers, especially those with security clearances, should be aware of these attempts, the memo said, adding that “if the offer seems flattering, urgent, exclusive or too good to be true, it probably is.”

A separate unclassified June 2024 transmission provided by the Army indicates the foreign intelligence activity has been observed for some time. It said adversaries use less traditional social messaging platforms like Reddit and Discord to pose seemingly innocuous questions to servicemembers that increase in sensitivity over time.

Requesters “may rely on opinion-based questions such as asking [Department of the Army] personnel for their opinion on topics such as Taiwan, Ukraine or Israel to gather information,” it said. The espionage efforts do not target only classified information and aim to access various forms of sensitive data like battle plans, contracts and research.

Army members and their families may also be extended “disproportionate payment” offers, such as $1,000 for a two-page whitepaper or an all-expense paid trip to destinations like China, Hong Kong or Macao, the 2024 message said.

A suspected Chinese intelligence operation has tried to recruit former U.S. federal employees and public policy experts through fake websites and job postings, Nextgov/FCW first reported in September.

Those recruitment efforts — which involved low-quality websites representing non-existent companies — appear to have picked up steam amid Department of Government Efficiency-fueled employee departures and terminations over the last several months.

The FBI previously told Nextgov/FCW that it’s aware of foreign adversaries using employment sites and social media platforms to identify knowledgeable individuals to target for recruitment.

“Often those targeted have security clearances and access to classified information. But our adversaries also are looking for experts in business and academia with technical expertise,” the bureau said.

Chinese intelligence entities have deployed online efforts to recruit unwitting current and former federal employees, the National Counterintelligence and Security Center said in April. In March, CNN reported that foreign adversaries, including China and Russia, accelerated efforts to recruit disgruntled federal workers in national security roles.

]]>

Industry groups push to keep open-source measures in annual intelligence bill

David DiMolfetta — Thu, 23 Oct 2025 14:21:00 -0400

A coalition of industry groups and former officials is asking Congress to preserve measures in the annual intelligence community authorization bill that support the use of open-source intelligence in U.S. spy agencies, according to letters first seen by Nextgov/FCW.

The missives are undersigned by firms like the Babel Street and the ANDECO Institute, which sell risk and threat intelligence services derived from commercially or publicly available information that’s not necessarily gathered through more covert means available to spy agencies. Graphika, which performs social media network analysis to identify disinformation campaigns, is also a signatory.

The measures, housed in Title 6 of the House Intelligence Committee’s version of the fiscal year 2026 Intelligence Authorization Act, are also supported by the OSINT Foundation, a professional association of open-source practitioners in the U.S. intelligence community. Former officials are also signatories, including Kristin Wood, who served as a deputy director in the CIA’s Open Source Center.

Open-source intelligence, dubbed OSINT, is the collection and analysis of publicly available data — like social media posts, news reports or satellite imagery — to generate insights that support national security and law enforcement decisions.

The Title 6 measures, in essence, aim to further legitimize the use of OSINT in day-to-day intelligence work, going as far as to require spy agencies to appoint senior officials responsible for managing and coordinating their OSINT activities.

“The arrival of mission-relevant artificial intelligence (AI) systems within the past two years means the [U.S. intelligence community] is, for the first time, able to make use of large unclassified datasets at scale to deliver decision advantage to U.S. policymakers and warfighters,” one of the letters says.

The House intelligence panel has zeroed in on OSINT this year, forming an open-source subcommittee. Last year, the Office of the Director of National Intelligence under then-director Avril Haines released an OSINT strategy, arguing that “rapid advances in artificial intelligence and machine learning present significant opportunities to capitalize on the value of OSINT.”

The missives are addressed to the Republican and Democratic leaders on the congressional intelligence panels. The sweeping intel bill is considered annually to authorize funding, programs and oversight of the U.S. spy community.

The letters come as the Senate and House are currently conferencing to resolve differences in the legislation. The Senate’s version does not have the same OSINT provisions in place as its lower chamber counterpart.

The letters aim to motivate both chambers to keep the House measures, though not all elements of the U.S. intelligence enterprise are on board. The National Geospatial-Intelligence Agency, which uses satellites and imagery analysis to track targets from space, is looking for a carve-out, according to Brandon McKee, the senior director for government affairs at the Special Competitive Studies Project’s Action Program, which backs the OSINT language.

A second person familiar with the matter, who requested anonymity because they were not authorized to speak publicly, also confirmed that NGA is not backing the measures at this time.

“They were not an entity that I believed would have been against a provision like this,” McKee said in an interview with Nextgov/FCW. “This is something that would greatly benefit the agency.”

Open-source intelligence can produce high-quality analysis for U.S. decisionmakers, often rivaling or complementing classified intelligence, while being faster to deliver and more affordable than other means, he argued. “If you’re looking to provide the best intelligence to policymakers and decisionmakers … this is another area that has to be there.”

There may be broader opposition. A Senate aide, who requested anonymity to speak candidly, said the Defense Department writ large is also opposed, as well as some other IC components. The aide did not name those agencies. The DOD houses NGA, as well as other intelligence giants like the National Security Agency and National Reconnaissance Office.

It’s not entirely clear why these spy agencies aren’t on board, though the hesitancy could stem from institutional concerns about losing control over their traditional mission space, budgets or legal authorities if open-source intelligence is elevated to the same level of legitimacy as classified work.

Nextgov/FCW has asked ODNI for comment.

“In the digital age, OSINT plays a critical role in real-time analysis, strategic warning and supporting tactical operations. We are proud that the HPSCI FY26 IAA, with the inclusion of the first-ever OSINT title, passed out of Committee overwhelmingly on a bipartisan basis,” a House Intelligence Committee spokesperson said. “Furthermore, we appreciate the letters of support from the nonprofit organizations and their recognition that the IC can more effectively and efficiently use open-source data.”

Spy agencies aren’t new to the use of OSINT, and many are predominant users of open-source means. Still, the acquisition and use of such data has been deemed controversial in the past.

Personal information on digital marketplaces like social media platforms is frequently packaged by data brokers, and spy agencies are among their customers. The dynamic has put the intelligence community on thin ice with some lawmakers and privacy advocates who call it an end-run around the Fourth Amendment, which bars unreasonable searches and seizures.

Last year, the Office of the Director of National Intelligence issued new policies that aim to guide spy agencies on best practices for ethically using commercial data. But agencies were not mandated to obtain a warrant before procuring or searching through datasets, a requirement that civil liberties groups have long advocated for.

OSINT has played a major role in recent armed conflicts, especially those in Gaza and Ukraine. Some groups argue that while open-source analysis has helped expose atrocities and verify battlefield developments, it also risks crossing ethical lines like revealing civilian locations, spreading unvetted or unverified data and operating in legal gray zones that aren’t immediately addressable under current humanitarian law.

]]>

The Netherlands joins US Air Force’s robot wingman program

Lauren C. Williams — Fri, 17 Oct 2025 16:54:00 -0400

WASHINGTON—The Netherlands wants in on the U.S. Air Force’s collaborative combat aircraft program to boost their own fleet of F-35 fighter jets. The Dutch Defense Ministry inked a letter of intent to cooperate in the program Thursday.

The Netherlands also signed a separate agreement with General Atomics as part of a broader effort to boost the country’s defenses and drone tech.

“We think that this is a unique point in time and it reinforces the partnership we have with the U.S. And I think it also makes the world a lot safer if in the near future we can actually also operate CCA type of aircraft in the European theater,” Gijs Tuinman, Dutch State Secretary for Defense, told reporters Thursday after announcing the agreement at the Dutch embassy’s defense industry event.

The country has partnered with the U.S. on the F-16 and F-35, which makes a CCA investment a natural next step that help proliferate the tech across Europe, Tuinman said, noting the Netherlands needs roughly equal numbers manned, unmanned, and attritable systems for its defenses.

“The Netherlands is like the jumping pad for the United States to get into Europe. So we have always [had a] strong or transatlantic relationship. That's my message here too: to sign the deal, but also to express that we understand the message from the U.S…that the Netherlands and Europe should shift the burden a bit” by increasing defense spending, Tuinman said.

The agreement allows the Netherlands access to the CCA program as it develops, to share data, and to provide input for requirements for use in Europe.

The Netherlands also penned an agreement with General Atomics to develop new small unmanned aircraft systems for intelligence, surveillance, and reconnaissance that are affordable and can hold a variety of payloads.

Tuinman said the drone industry lacks systems that can “penetrate [anti-access/area-denial] bubbles and have a diverse set of ISR and strike capabilities.”

General Atomics will work with Netherlands-based VDL Defentec to engineer and produce the new systems.

The move comes months after General Atomics and fellow CCA-maker Anduril began pitching tailorable versions of the platform—and co-production—to European countries at the Paris Air Show this summer.

The Dutch partnership aims to address immediate security threats as Russia’s war on Ukraine persists and drone activity increases across Europe. Drones recently disrupted communications during a Dutch military exercise in Poland.

“Putin is testing us in every possible way,” Tuinman said. “Hybrid attacks are already taking place across Europe…including my own country. And over the past weeks, various locations in Europe have been plagued by large amounts of mysterious drones testing the strength of our response, resilience, and most of all our alliance.”

Teaming with General Atomics, and other U.S. defense companies, also creates an opportunity to bolster defense industries on both sides of the Atlantic Ocean, Birgitta Tazelaar, the Dutch ambassador to the U.S., said Thursday during opening remarks at the embassy’s annual defense industry event.

Spending more on defense “means that we're going to build up a European defense industry, but it also means that we're going to work very well together with our American partners in doing so. And this is crucial. Look at our adversaries and our competitors. They are doing the same, and it is extremely important to keep our strategic advantage by working together and integrating our industrial bases to the extent that we both benefit,” Tazelaar said.

]]>

DOD would curtail long-term intelligence work during government shutdown

David DiMolfetta — Tue, 30 Sep 2025 12:45:00 -0400

The nation’s top spy offices are expected to pare down certain intelligence-gathering activities deemed non-essential in the event of a government shutdown that is anticipated at midnight Tuesday.

Under shutdown guidance provided by the Defense Department, intelligence work that directly supports active military operations, threat monitoring or other national security emergencies is designated “excepted” and would continue uninterrupted if funding lapses.

But agencies would be required to pause certain longer-term activities. Those include political and economic analysis work unrelated to current crises and intelligence support for weapons acquisition.

Political and economic assessments can assist military planners’ understanding of how foreign governments and global financial conditions shape conflicts, while weapons acquisition intelligence helps the U.S. design, test and purchase systems that can survive against current or emerging threats.

In essence, tactical intelligence collection activities would remain active, though much of the strategic analysis that supports future planning of the DOD’s spying activities would be curtailed until federal funding is restored.

“Command, control, communications, computer, intelligence, surveillance and reconnaissance activities” remain excepted functions, the document says. That also includes the use of spying capabilities tied to telecommunications infrastructure, which are often used by the National Security Agency to intercept phone calls and other communications as they cross the world’s internet backbone.

Offices like the National Geospatial-Intelligence Agency, which relies on satellites and imagery analysis to track targets from space, can also continue their core intelligence missions. Other major DOD spying offices include the Defense Intelligence Agency and the National Reconnaissance Office, the latter of which designs and launches the nation’s spy satellites.

The exemptions would also apply to a slew of other intelligence units housed inside military branches like the Army, Air Force and Navy.

Other intelligence offices like the CIA are not housed directly in DOD but coordinate closely with the military on spying matters. Less public information is available on shutdown plans for the CIA and the Office of the Director of National Intelligence, which oversees the nation’s 18 spy agencies.

]]>

Pentagon bans tech vendors from using China-based personnel following a ProPublica investigation

Renee Dudley, ProPublica — Mon, 22 Sep 2025 15:43:00 -0400

What Happened

The Defense Department has tightened cybersecurity requirements for tech companies that sell cloud computing services to the Pentagon.

The updates, issued this month, ban IT vendors from using China-based personnel to work on department computer systems and require companies to maintain a digital paper trail of maintenance performed by their foreign engineers.

Background

The changes follow a ProPublica investigation that exposed how Microsoft used China-based engineers to maintain government computer systems for nearly a decade — a practice that left some of the country’s most sensitive data vulnerable to hacking from its leading cyber adversary.

U.S.-based supervisors, known as “digital escorts,” were supposed to serve as a check on these foreign employees, but we found they often lacked the expertise needed to effectively supervise engineers with far more advanced technical skills.

What They Said

The Defense Department now says in its “Security Requirements Guide” that only “personnel from non-adversarial countries” may work on its cloud systems and that the escorts supervising those foreign workers “must be technically qualified in the code/system or technology they are providing access to.”

In addition, cloud providers must maintain detailed audit logs, a digital trail of actions in computer systems. The logs “must include identification of the escort and escorted,” including country of origin, as well as details of commands executed and settings changed.

Why It Matters

Until our reporting, top Pentagon officials said they had been unaware of Microsoft’s digital escort system, which the company developed as a work-around to a Defense Department requirement that people handling sensitive data be U.S. citizens or permanent residents.

Cybersecurity and intelligence experts have told ProPublica that the arrangement poses major risks to national security, given that laws in China grant the country’s officials broad authority to collect data. Leading members of Congress, in turn, have called on the Defense Department to strengthen its security requirements while blasting Microsoft for what some Republicans called “a national betrayal.”

The Pentagon is now conducting an investigation into the digital escort program, with a focus on Microsoft’s China-based engineers.

Response

Following ProPublica’s reporting, Microsoft announced in July that it would stop using China-based engineers to service Defense Department cloud systems. In a statement for this article, a spokesperson said the company was committed to implementing the department’s new requirements.

“Our commitment to national security is foundational, and we remain focused on providing the most secure services possible to the US government,” the spokesperson said. “We recently implemented changes to our Department support model, and will continue to work with our national security partners to evaluate and adjust our security protocols in light of the new directives.”

Doris Burke contributed research. ProPublica is a nonprofit newsroom that investigates abuses of power.

]]>

Pentagon sets start date for CMMC implementation

Ross Wilkers — Wed, 10 Sep 2025 07:47:46 -0400

Circle Nov. 10 as when the Defense Department’s new cyber and supply chain security standard for the entire industrial base starts to be implemented, almost six years after Pentagon leadership began talking about it.

The final rule for the Cybersecurity Maturity Model Certification 2.0 standard went into effect in December 2024, while the next step to implement the program into contracts started Tuesday with a regulation released for public inspection.

Also called the 48 CFR rule, its publication amends the Defense Federal Acquisition Regulation Supplement that governs all Pentagon contracts. Industry now has two months’ notice before CMMC 2.0 begins to appear in DOD solicitations.

CMMC 2.0 is the Pentagon’s new set of requirements for companies that house controlled unclassified information or federal contract information in their systems. Companies have up to three levels of compliance they can be certified under, depending on how sensitive the information is.

DOD plans to roll out the program in a four-phase process over the next three years. In the first phase starting Nov. 10, solicitations will require self-assessments at certification Levels 1 and 2 where applicable.

Some Level 2 certifications will require a verification check done by a certified third-party assessor organization if the data is considered more sensitive. Any and all Level 3 applicants will require certification from the Defense Industrial Base Cybersecurity Assessment Center.

The release of the new 48 CFR rule indicates industry can also circle Nov. 10, 2028, as when all DOD solicitations and contracts will mandate some level CMMC compliance for eligibility to bid for the work.

Stay in the know — Washington Technology’s Insider Membership gives you unmatched access to breaking news, in-depth analysis, and insights that federal contractors can’t afford to miss. Join today for 50% off.

]]>

Pentagon stands up new task force to coordinate anti-drone efforts

Meghann Myers — Fri, 29 Aug 2025 15:47:00 -0400

Five years after the Pentagon created an office to coordinate its counter-drone efforts, it’s trying again.

The new Joint Interagency Task Force 401 will spearhead the acquisition and integration of air defense systems to take down small unmanned aerial systems, Defense Secretary Pete Hegseth announced Thursday in a video first posted to X.

The group will “rapidly deliver Joint C-sUAS capabilities to America's warfighters, defeat adversary threats, and promote sovereignty over national airspace,” Hegseth wrote in a memo dated Wednesday.

The memo also shuts down the existing Joint Counter-small Unmanned Aircraft Systems Office, which has been around since February 2020. It will also integrate the department’s Replicator 2 efforts.

“The JCO had great intentions but struggled to compel the different services and organizations to participate,” an Army official, who was not authorized to speak on the record, told Defense One. “Whereas the JIATF will have a lot more ability to coordinate and compel.”

Hegseth’s “priorities for transformation and acquisition reform include improving C-sUAS mobility and affordability and integrating capabilities into warfighter formations,” he wrote in the memo.

Hegseth’s memo lays out several guidelines for standing up the new task force:

It will have a director with acquisition authority, who will submit unfunded requirements for the 2026 fiscal year within the next 30 days.
It will immediately begin recruiting a technical lead and four personnel from the military services “with operations, acquisition, electronic warfare (EW), intelligence, or other C-sUAS competencies to include one officer in the grade of O-5 or higher who will have access to his or her Military Service' s decision-making officials.”
The under secretary of defense for research and engineering has 30 days to make recommendations on establishing a designated c-sUAS test and training range.
The Army has five days to submit requirements to the Pentagon building’s management for office space required to house the JIATF.
The Army has 30 days to submit its full implementation plan to the defense secretary, and will update the secretary on progress monthly.

In addition to this new DOD-wide effort to procure and employ counter-drone capabilities, the services have been working on their own acquisitions, which will continue, according to the memo.

The Marine Corps this summer began fielding counter-drone systems to every infantry squad, while the Army is working on its own solutions.

The JIATF, the Army official said, can coordinate some of those efforts across the services.

“If we’ve got a good solution to a problem that everybody has, let’s scale that solution, vice everybody trying to solve the problem independently,” the official said.

]]>

Defense tech office cuts staff down to 40

Alexandra Kelley — Mon, 04 Aug 2025 16:39:00 -0400

The Department of Defense announced staff cuts within the agency’s Defense Technical Information Center — which it initially announced as the Defense Technology Information Center — on Monday, marking another in a series of personnel reductions under the second Trump administration.

Sent out via an email, a statement from Chief Pentagon Spokesman and Senior Advisor Sean Parnell said that the decision will eliminate duplicative functions and roles with a goal of refocusing DTIC to “its core statutory mission of administering a library of technical information and improving the user experience.”

Parnell said that this reduction in force, directed by the agency’s Under Secretary for Research and Engineering, is set to save $25 million dollars in federal spending.

A total of 40 personnel will be retained at the office, having been identified as mission-essential, and all other civilians will receive notices of reduction-in-force by August 25. That marks a nearly 80% reduction from the 193 employees working in that office as of September 2024, per Office of Personnel Management employment data. The office, according to its website, is responsible for sharing DOD’s science and technology investment with all of the services, “enabling lab scientists, engineers, and researchers to build upon past and present research.”

The reduction-in-force follows the Pentagon’s announcement of a similar downsizing within the Office of the Director, Operational Test and Evaluation in May. Defense Secretary Pete Hegseth said at the time that the removals, which reduced the staff at DOT&E from 94 employees down to 30 civilians and 15 service members, was a result of a “comprehensive internal review” that identified duplicative roles and positions in the office.

In February, roughly 61,000 Defense employees’ jobs were terminated, beginning with probationary employees, as part of the Department of Government Efficiency’s broad cuts across the federal government to restructure agency operations and reduce federal spending.

]]>

The Army’s giant data deal with Palantir is a harbinger: service CIO

Lauren C. Williams — Mon, 04 Aug 2025 08:22:19 -0400

The Army collapsed 75 data and software contracts into one deal potentially worth $10 billion—a deal won by Palantir—and it’s just the beginning of how the service wants to buy software.

“This is going to be one of many enterprise licensing agreements that we're looking at entering into,” Army CIO Leonel Garciga told reporters.

The goal is to reduce costs, while making it easier to buy software and getting the functions the service needs “on demand.”

“We have a lot of big software packages that are out there. They've been bought over several years, several program offices, several commands [and] not getting a lot of parity across the board on how they're being delivered. Adding a lot of complexities,” Garciga said. “But our intent is to continue to move down this path, to really focus on reducing that complexity, adding agility to how we buy….[and] save taxpayer dollars as much as we can.”

The contract announced late Thursday has a $10 billion cap with a 10-year performance period and allows the Army and, potentially, other Defense Department agencies, to buy Palantir products, including AI tools and data analytics. It also deepens the company’s already close relationship with the Army and Pentagon.

“We've got a plethora of Palantir employment inside the Army—everything from Army intel data platform to Vantage to much smaller projects…that we're paying for that we would be using this vehicle to also procure,” Garciga said.

The Army is working with other vendors on similar contracts, he said.

The service has been working to simplify its IT contracts and streamline how it buys software in recent years. But the Palantir award and new approach comes after Defense Secretary Pete Hegseth called for new software acquisition practices earlier this year.

"We have been working on this since November of last year, and I think that there was just an inherent understanding, you know, almost two years ago now, that we needed to start moving in this direction with a handful of our vendors,” Garciga said.

Still, he said, the secretary’s memo has been a “catalyst” for some “commercial partners to rethink the way that they integrate and work with us in the government, and what our contractual agreements are going to look like moving forward.”

Danielle Moyer, the executive director at Army Contracting Command, said it’s part of a “common-sense” effort to make sure the Army doesn’t have duplicative contracts.

“If I have more than one contract with the same vendor, have I bought the same thing more than once in a different way or at a different price?” Moyer said.

The effort, which starts with this giant award to Palantir, is to “make sure we're getting the best discount.”

]]>