Anthropic, Google and xAI will join OpenAI on the CDAO’s nascent effort to partner with industry on pioneering artificial intelligence projects focused on national security applications. Under the individual contracts — each worth up to $200 million — the Pentagon will have access to some of the most advanced AI capabilities developed by the four companies, including large language models, agentic AI workflows, cloud-based infrastructure and more.

“The adoption of AI is transforming the Department’s ability to support our warfighters and maintain strategic advantage over our adversaries,” Chief Digital and AI Officer Doug Matty said in a statement. “Leveraging commercially available solutions into an integrated capabilities approach will accelerate the use of advanced AI as part of our Joint mission essential tasks in our warfighting domain as well as intelligence, business, and enterprise information systems.”

OpenAI received the first contract for the effort June 17 and will create prototypes of agentic workflows for national security missions. According to CDAO, work with all four vendors will expand the Pentagon’s experience with emerging AI capabilities, as well as give the companies better insights into how their technology can benefit the department.

The contract with CDAO is also another win for xAI, which is owned by Musk — who previously led the Trump administration’s Department of Government Efficiency (DOGE) efforts but recently had a falling out with President Donald Trump over legislation and other issues — and develops the generative AI tool called Grok. The company announced Monday that it was launching a new suite of AI tools for U.S. government users known as “Grok for Government.” The platform is now available to purchase by federal agencies through the General Services Administration, according to a post on X, which Musk also owns.

In a blog post published Monday, Jim Kelly, Google Public Sector’s vice president of federal sales, noted that the company will provide the Pentagon its Cloud Tensor Processing Units for training AI models, AI-powered agents via Google’s Agentspace, and access to the company’s infrastructure based in the contiguous United States.

“These advanced AI solutions will enable the DoD to effectively address defense challenges and scale the adoption of agentic AI across enterprise systems to drive innovation and efficiency with agile, proven technology,” Kelly wrote.

The announcement is the latest step the Defense Department has taken in recent months to accelerate adoption of AI-enabled capabilities developed by commercial companies — many of which have recently announced new business ventures focused on national security.

In June, Anthropic introduced a custom set of its Claude Gov AI models that are tailored specifically to defense use cases, ranging from operational planning to intelligence analysis. The same month, OpenAI launched a new initiative called “OpenAI for Government” that expands on its current partnerships with the Defense Department and other U.S. government agencies — including custom AI models for national security.

The post Pentagon awards mega contracts to Musk-owned company, other firms for new ‘frontier AI’ projects appeared first on DefenseScoop.

In its daily list of new contract awards, the department announced Monday evening that the CDAO had awarded a $200 million prototype other transaction agreement to OpenAI Public Sector to “develop prototype frontier AI capabilities to address critical national security challenges in both warfighting and enterprise domains.”

The estimated completion date for the work is July 2026.

The announcement did not provide any additional details about what those capabilities will entail or what specific mission sets they’ll be applied to.

Last year, the CDAO listed a variety of “warfighting” use cases that its newly launched Artificial Intelligence Rapid Capabilities Cell would focus on, including command and control, decision support, operational planning, logistics, weapons development and testing, uncrewed and autonomous systems, intelligence activities, information operations and cyber operations.

“Enterprise management” use cases include financial systems, human resources, enterprise logistics and supply chain, health care information management, legal analysis and compliance, procurement processes, and software development and cybersecurity.

On Tuesday, DefenseScoop sent the CDAO questions seeking more information about the deal with OpenAI.

In a statement, a CDAO official said the vendor will be expected to “prototype agentic workflows to address our hardest challenges.”

The official, who provided comment on the condition of anonymity, added that more announcements are expected in the near term.

“This award is another step on our journey toward accelerating the adoption of advanced AI capabilities across the Department, and, in the coming weeks, we will announce partnerships with other Frontier AI companies as well. Access to top tier talent from partners like OpenAI is critical for building the agentic workflows needed to increase Joint Force lethality and enterprise efficiencies,” they said.

They noted that the CDAO has recently partnered with the Army’s Enterprise Large Language Model (LLM) Workspace to provide military users across the combatant commands, Office of the Secretary of Defense, and the Joint Staff access to “industry-leading general purpose LLMs.”

In an press release Monday evening announcing the launch of a new “OpenAI for Government” initiative, the company noted the new $200 million deal with the CDAO, saying the organization will help the Pentagon “identify and prototype how frontier AI can transform its administrative operations, from improving how service members and their families get health care, to streamlining how they look at program and acquisition data, to supporting proactive cyber defense.”

More broadly, the company said the OpenAI for Government initiative will give customers access to its “most capable” models within “secure and compliant environments,” including through ChatGPT Enterprise and ChatGPT Gov.

It will also offer custom models for national security organizations on a limited basis, per the announcement.

Katrina Mulligan, the head of the company’s national security policy and partnerships who previously served as chief of staff to the secretary of the Army and principal deputy assistant secretary of defense for special operations and low-intensity conflict, told DefenseScoop that the new project with the CDAO marks a major milestone for OpenAI for Government as it moves to expand its work in the public sector.

“The $200 million ceiling reflects the Department’s trust in OpenAI to responsibly bring frontier AI into mission-critical settings. CDAO is also creating a framework for how OpenAI’s leading expertise can be used to develop and test solutions that support the Department of Defense in its mission to ensure the safety and security of Americans,” she wrote.

An OpenAI spokesperson said the goal of the pilot program is to “scope” and prototype potential applications for frontier AI.

“The contract aims to explore where OpenAI’s tools can improve military operations and cybersecurity, save time for staff by making tedious work more efficient, and help DoD better support service members. DoD requires this type of contract to scope future partnerships, and the project is structured to lead into a potential follow-on production agreement. The applications may take different functions and forms that are yet to be determined – but could include all of our services,” they told DefenseScoop, adding that all use cases must be consistent with the company’s usage policies and guidelines, which “prohibit its use for the development or use of weapons.”

The announcement of the $200 million deal came just a few days after OpenAI’s chief product officer Kevin Weil joined the Army Reserve to serve in Detachment 201, a new “Executive Innovation Corps,” along with other execs from the tech community.

Mikayla Easley contributed reporting for this story.

The post Pentagon tapping OpenAI, other vendors for ‘frontier AI’ projects appeared first on DefenseScoop.

The move is the latest push by the department to tap into capabilities and know-how from Silicon Valley and the commercial sector.

The new corps “brings top tech talent into the Army Reserve to bridge the commercial-military tech gap” and is “designed to fuse cutting-edge tech expertise with military innovation,” the Army stated in a press release.

On Friday, the service is set to swear-in Meta’s chief technology officer Andrew Bosworth, OpenAI’s chief product officer Kevin Weil, Palantir’s CTO Shyam Sankar and Bob McGrew, an advisor at Thinking Machines Lab who was previously OpenAI’s chief research officer.

Meta, which owns Facebook, recently announced a new partnership with defense tech company Anduril to develop extended reality (XR) products for soldiers.

OpenAI is the maker of the wildly popular generative AI tool ChatGPT. The Army and the Defense Department writ large are pursuing new genAI tools to boost productivity and efficiency.

Palantir is a major provider of software tools for the DOD — including the Maven Smart System — and is also developing hardware, such as the Army’s AI-enabled TITAN vehicle.

“Det. 201 is an effort to recruit senior tech executives to serve part-time in the Army Reserve as senior advisors. In this role they will work on targeted projects to help guide rapid and scalable tech solutions to complex problems. By bringing private-sector know-how into uniform, Det. 201 is supercharging efforts like the Army Transformation Initiative, which aims to make the force leaner, smarter, and more lethal,” the service stated in Friday’s press release.

The swearing-in of the four new officers “is just the start of a bigger mission to inspire more tech pros to serve without leaving their careers, showing the next generation how to make a difference in uniform,” per the release.

The Army didn’t provide additional details about how large the detachment will grow to or how fast the service will expand it by bringing in new personnel from the tech sector.

Friday’s announcement comes in the midst of a new Army Transformation Initiative that was launched in recent weeks — which is being spearheaded by Secretary Daniel Driscoll and Chief of Staff Gen. Randy George — that calls for eliminating systems that are deemed obsolete for soldiers on the battlefield in the future and procuring “dual-use” capabilities. Driscoll has advocated for buying more commercial off-the-shelf tech, among other reforms.

The post Army recruits officers from Meta, OpenAI and Palantir to serve in new detachment appeared first on DefenseScoop.

In 2023, Bonnell was tapped to serve as AFRL’s first-ever chief information officer and director of the laboratory’s Digital Capabilities Directorate, where she led the lab’s information technology strategy and overall modernization efforts. According to a Tuesday post on LinkedIn, Bonnell is now working at OpenAI as a partnership manager, a position she took on in March.

“The role [at AFRL] was truly one in a lifetime — serving the national security mission alongside some of the most brilliant scientists, engineers, and digital visionaries in the country,” Bonnell wrote. “From cybersecurity to networks and infrastructure, from enterprise service design to pushing the frontiers of AI, I couldn’t be more proud of what we built together—or more confident in the team carrying the mission forward.”

While at AFRL, Bonnell was at the forefront of the lab’s push to develop new artificial intelligence capabilities for warfighters. She was instrumental in launching the Air Force’s experimental generative AI chatbot known as NIPRGPT — a model that has since been scaled to other organizations across the Defense Department such as the Defense Information Systems Agency.

“Helping to launch one of the first human-machine teaming research platforms in DoD, built with open-source tools and volunteer effort, was a career highlight. So was advancing AI adoption across the force, making the theoretical practical,” Bonnell wrote.

Before joining AFRL, Bonnell was Google Public Sector’s emerging tech “evangelist,” where she helped the Defense Department and other federal agencies adopt new capabilities such as cloud, AI and zero-trust cybersecurity strategies.

As she returns to the private sector, she expects to dive deeper into artificial intelligence capabilities and explore how the technology can contribute to public service, she said on LinkedIn.

At OpenAI, Bonnell will “support extraordinary public sector organizations like the U.S. National Labs to research and apply frontier current and future AI models to the grand challenges of research, science, innovation, and national security,” she wrote.

The post Former AFRL CIO, director of digital capabilities joins OpenAI appeared first on DefenseScoop.

Microsoft announced Thursday it received authorization for 26 additional products in its top-secret cloud environment, meeting Intelligence Community Directive (ICD) 503 standards and allowing agencies — particularly those in the intelligence community and Defense Department — to use them for the government’s most classified information. Those added tools include Azure OpenAI Service — which provides Azure customers access to OpenAI’s generative AI large language models — and Azure Machine Learning, among others.

Douglas Phillips, Microsoft corporate vice president, wrote a blog post announcing the news that Azure OpenAI “allows agencies and authorized partners operating in Microsoft’s Azure Government Top Secret cloud to benefit from multimodal generative AI models, such as GPT-4o, while meeting the rigorous security and compliance requirements necessary for the nation’s most sensitive data. Authorized users can easily access and integrate Azure OpenAI Service and further ground it on their data for more specialized and accurate intelligence.”

GPT-4o is an OpenAI model that can be used for natural language understanding and processing, text summarization and classification, sentiment analysis, question answering, conversational agents and more. It is the foundational model that the popular commercial generative AI tool ChatGPT is built on.

The announcement comes after Azure OpenAI received FedRAMP High authorization last August. 

Last May, William Chappell, Microsoft’s chief technology officer for strategic missions and technologies, told DefenseScoop that the company  had deployed OpenAI’s GPT-4 to an isolated, air-gapped Azure Government Top Secret cloud for use by the Department of Defense for testing. However, the model wasn’t accredited for wider use at the time. The accreditation announced Thursday would now make that possible.

Chappell told DefenseScoop the availability of GPT-4 in the top-secret environment would help DOD officials deal with vast amounts of data.

It’s about “making sure you have the right information at the right time,” he said. “So whether it’s geospatial or any amount of data, we’re swimming in data, we’ve got sensors everywhere. How do you actually make sense of the information that is within your organization? Whether that’s proposals or all sorts of different types of paperwork that we all have to do — how do you simplify and how do you sort through that … data that’s mission focused or data that’s more back office and human resource-focused?”

The post OpenAI’s GPT-4o gets green light for top secret use in Microsoft’s Azure cloud appeared first on DefenseScoop.

Once the tool is accredited, Pentagon officials will be able to use the technology in a secure environment.

“When you’re connected to the internet, you’re always having to defend against people probing and trying to wreak havoc. And so now you’ve got an isolated and an air-gapped environment …. Now we have the capability that we have in the unclassified side, in that [top-secret] environment for the very first time. So all of the tools … you see around here, people using this, is now available for those type of [top-secret] uses,” William Chappell, chief technology officer for strategic missions and technologies, told DefenseScoop during an interview on the sidelines of an AI expo hosted by the Special Competitive Studies Project, where he announced the deployment.

The technology will help DOD officials deal with vast amounts of data, he said.

It’s about “making sure you have the right information at the right time. So whether it’s geospatial or any amount of data, we’re swimming in data, we’ve got sensors everywhere. How do you actually make sense of the information that is within your organization? Whether that’s proposals or all sorts of different types of paperwork that we all have to do — how do you simplify and how you sort through that … data that’s mission focused or data that’s more back office and human resource-focused?”

Microsoft is a major investor in OpenAI, the maker of GPT-4 and the popular ChatGPT.

Pentagon officials aim to leverage generative AI capabilities such as large language models, but they want to acquire solutions that won’t expose classified information to unauthorized individuals. They also want capabilities that can be tailored to meet DOD’s unique needs.

With the GPT-4 capability that Microsoft has deployed via Azure OpenAI Service in Azure Government Top Secret, “they can fine tune it, they can add their own data, they can do a bunch of things… They’ll build unique workflows on top of that, specific to their mission,” Chappell said.

“OpenAI is, in my opinion, best in class. And then we built security Copilot and we built M 365 Copilot … That intelligence that’s under the hood is then picked up out of the commercial domain, out of our product suite, and then over so that the government can build those similar type of applications,” he told DefenseScoop.

He declined to predict when the GPT-4 tool will be accredited for top-secret work.

“I don’t want to say an exact date because the government has a say. Right? So we’ll be working hand-in-hand with the government from this moment, now that it will actually write code and give you information, to how we harness stuff, how’s it deployed, you know, the right way and make sure it’s presented to the end user the right way. That’s part of the accreditation process that they really control,” he said.

The post Microsoft deploys GPT-4 large language model for Pentagon use in top secret cloud appeared first on DefenseScoop.

Generative artificial intelligence has made headlines recently with the emergence of OpenAI’s ChatGPT technology and other capabilities that can generate new content, such as text or images, based on their training using large language models or other tools.

During an event hosted by the Center for Strategic and International Studies on Tuesday, NSA cybersecurity director Rob Joyce was asked if technology like ChatGPT will help hackers create more effective phishing messages.

“I absolutely believe that,” Joyce said. “People have gone across the scale of, you know, how worried should they be about ChatGPT? I will tell you the technology’s impressive, right. It is really sophisticated. Is it going to, in the next year, automate all of the attacks on organizations? Can you give it a piece of software and find tell it to find all the zero-day exploits for it? No. But what it will do is it’s going to optimize the workflow. It’s going to really improve the ability for malicious actors who use those tools to be better or faster.”

“And in the case of the malicious foreign actors, it will craft very believable native language, English text that could be part of your phishing campaign or your interaction with a person or your ability to build a backstory — all the things that will allow you to do those activities or even malign influence, right. That’s going to be a problem. So, is it going to replace hackers and be this super AI hacking [tool]? Certainly not in the near term, but it will make the hackers that use AI much more effective, and they will operate better than those who don’t,” he added.

There’s also a risk that foreign competitors might also try to steal the intellectual property behind some of these generative AI technologies.

“I can’t talk to any specific threats. But … all of our industrial advancements that are game-changing have been targeted in the past, right. Whether it’s material science or chemicals or battery technology, I don’t care what it is — if we’ve innovated it and have the state of the art, you know it’s been under pressure from China and others to pull that and steal and bypass the investments our companies are making to develop it. And so I see no reason that there’s not a major focus on getting those [AI] models and bypassing all the investment and, you know, the capital it took to develop them,” Joyce said.

However, while generative artificial intelligence poses a threat, officials also believe it could be a useful tool for the U.S. intelligence community. The NSA and CIA, for example, are expressing interest in these types of capabilities.

However, the technology currently has major shortcomings, experts say.

“I don’t know how many of you have played with … name your favorite model out there, but they hallucinate. And that’s the technical term of art meaning they will generate data that’s not real. I have to be able to generate real data to bring it to a company or the president [of the United States] or the warfighters. So, I have to get to the point where we’re understanding that outputs are factually accurate. In my world, that’s a high bar,” Joyce said.

“The idea that it will sort and provide acceleration — just like I talked about the advantage to the adversary — you know, every single day our analysts are overwhelmed with … what they have to focus on. If it can raise things up and even if it’s only 90% right on the things it surfaces, then the human can work in a much more enriched flow. And at that point, they become more effective,” he added. “So, I think that’s … the sweet spot. It’s a tool, [but] that’s not going to replace our analysts.”

Leaders at OpenAI have acknowledged some of the risks associated with the technology, including the threat of hallucinations and cyberattacks, while also touting its benefits.

The Pentagon is eyeing generative AI as a tool that could aid the military with “decision support and superiority.”

The Defense Department is scheduled to host a conference in McClean, Virginia, in June to discuss the technology. The confab is still in the early stages of planning, a DOD spokesperson told DefenseScoop recently.

However, Pentagon officials are also concerned about the hallucination problems that Joyce mentioned on Tuesday.

“On the trusted AI and autonomy front, we are working with companies that are developing these large language models. And we recognize that, as all you should recognize that these things don’t have any common sense and they are word completion programs that use statistics to complete thoughts, you know, based on prompts,” Deputy CTO for Critical Technologies Maynard Holliday said at the Potomac Officers Club’s annual R&D Summit last month.

“It still hallucinates greatly and it does not have a corpus of defense-specific information that would give us decision advantage yet. And so, you know, we recognize that we’re going to have to develop that corpus of data so that these large language models can be useful — and then to minimize, you know, the hallucinatory side effects of what these logic language models do,” he added.

The post NSA: ChatGPT and similar tech will make hackers more effective appeared first on DefenseScoop.