Dataminr Pentagon contract: $318M AI alerting, every branch

The Dataminr Pentagon contract, worth roughly $318 million over five years, puts AI-powered alerting on publicly available information at the center of U.S. military early warning. The firm-fixed-price award, announced this week under the A2 Publicly Available Information Alerting program, spans every armed service branch and covers three mission areas: monitoring global events, protecting deployed forces, and issuing advance warning of developments that could affect U.S. forces.

The Pentagon ran the procurement as a competition and said it received eight bids. Work will be performed in Washington, D.C., with completion expected in June 2031. At award time, the department obligated $46.5 million in fiscal 2026 operations and maintenance funds, and the balance of the $317.8 million ceiling will be drawn down over the life of the deal.

The contract structure carries a clue about how the program will run. Because the award is firm-fixed-price, cost risk sits with Dataminr rather than the government. And since only $46.5 million was obligated up front, about 15 percent of the ceiling, actual spending will track annual appropriations across the next five fiscal years: the ceiling is a ceiling, not a guarantee.

Inside the Dataminr Pentagon contract

The requirement behind the A2 program originated with Headquarters Air Force's Deputy Chief of Staff for Intelligence, Surveillance, Reconnaissance, and Cyber Effects Operations, which called for a commercially available platform able to supply near-real-time warning from open sources. Contracting runs through the Air Force District of Washington Directorate at Joint Base Andrews, Maryland, even as the capability expands well beyond the Air Force.

Funding the work out of operations and maintenance accounts is itself a signal. The program is being treated as an operational capability, which is why it can be scaled across the whole department. Every service branch, from the Army and Navy to the Marine Corps, Air Force, and Space Force, sits in scope, along with all unified combatant and operational commands.

What the A2 platform does

Dataminr, which describes itself as a native AI company, built the platform on a stack of language models. More than 60 of them run around the clock, each tuned to detect, verify, and interpret events the moment they surface in public material: government statements, defense publications, procurement disclosures, sensor data, imagery, and video, across more than 150 languages and over a million public sources.

The models go beyond flagging that something happened. Specialized LLMs compare emerging events against historical patterns to forecast how a situation is likely to evolve and what its impact may be, giving commanders an early read on developments that could affect deployed forces. That predictive layer is what separates the service from a plain news-monitoring feed; the system is built to estimate trajectory and downstream effects. The same open-source engine also contributes to space domain awareness, tracking public signals relevant to orbital activity so satellite operators can corroborate sensor data.

After an event passes detection and validation, alerts go out immediately, so force-protection and intelligence decisions can keep pace with fast-moving situations. Dataminr's pitch is that public signals are matched and reported before the same information reaches conventional reporting channels.

First Alert Advanced and the Palantir integration

First Alert is not new to the Pentagon. The Air Force has been the issuing service for the program, and Dataminr says it created the real-time public-data alerting category it now serves. The new award formalizes that relationship, extends it through June 2031, and widens it from a single service to the entire department.

Alongside the award, Dataminr launched First Alert Advanced, an agentic, autonomous real-time intelligence platform for defense and government customers. Where earlier First Alert generations surfaced alerts for analysts to triage, the new tier adds agentic correlation and follow-through, letting the system assemble context, prioritize warnings, and route them into operational workflows. The product launch and the contract were announced together, pairing new capability with the funding to deploy it.

A new integration with Palantir Foundry is central to the offering. It lets open-source intelligence from Dataminr be combined with internal government data layers, so a public-signal alert can be checked against classified or operational holdings inside a single workflow. The timing shows how the Pentagon's AI vendor map is converging: a day before the Dataminr award, the Army expanded Palantir's Maven Smart System contract by $618 million, which the service described as the largest AI software contract in Army history, layered on the $10 billion, 10-year enterprise agreement Palantir signed in July 2025.

Read together, the Dataminr Pentagon contract and the Palantir expansion describe a division of labor. Dataminr supplies the open-source detection layer; Palantir supplies the fusion substrate where that intelligence meets government data; the services supply the operational context that makes the alerts actionable.

The wider defense AI wave

The Dataminr Pentagon contract landed in a busy week for defense AI procurement. Leidos took a $301 million, five-year follow-on to run 24/7 cyber operations and develop AI-enabled resilience capabilities for Army networks, and Safe Pro Group booked a $780,000 order for AI-powered drone threat detection and uncrewed systems. The pattern is consistent: the Pentagon is spending across sensing, fusion, and alerting instead of concentrating on single flagship programs.

For Dataminr itself, the award locks in a predictable federal revenue stream at a moment when a public listing remains an open question for the company. Dataminr framed the deal as cementing its position as the military's real-time alerting backbone for the next half-decade, and the five-year term gives that claim a contractual footing. For rivals in the public-data intelligence market, the award closes off the largest available slot in the category until 2031.

There is also a governance dimension to the Dataminr Pentagon contract. Dataminr's public-data alerting previously found a market with domestic police departments, where its use in protest monitoring drew criticism over the absence of oversight. The A2 program extends the same class of technology into military early warning, and the oversight questions that accompanied the law-enforcement deployments do not automatically disappear when the customer becomes the military.

Why this matters

For defense decision-makers, open-source intelligence just became a first-class operational input rather than an auxiliary feed. The award also shows how quickly the Pentagon's AI vendor map is consolidating, with Dataminr's detection layer and Palantir's fusion layer effectively becoming complementary defaults. For Dataminr, the five-year revenue base and the Foundry partnership create a position in the military AI stack that competitors will find hard to dislodge.

Related Articles

• Pentagon memo lines up $244M in no-bid Palantir contract work through 2028
• Palantir's Maven Program of Record Status Locks In $2.3B in Pentagon Funding
• Navy Taps Domino Data Lab for $99.7 Million AI Mine Detection Initiative

✔Human Verified

Researched and cross-referenced against primary sources by the Bytevyte editorial team. This article was generated with the assistance of artificial intelligence and reviewed by the Bytevyte editorial team.