- Updated: February 25, 2026
- 6 min read
Pentagon AI Advisory Team Gains Industry Heavyweights Emil Michael and Steve Feinberg
The Pentagon’s new AI advisory team, led by former Uber executive Emil Michael and private‑equity billionaire Steve Feinberg, is shaping U.S. defense AI policy, confronting supply‑chain risks, and redefining how the military will deploy frontier AI models.
Pentagon AI Advisory Team: Emil Michael, Steve Feinberg, and the Future of U.S. Defense AI
Quick Overview
The Department of Defense has assembled a high‑profile AI advisory group to steer its defense AI strategy. The team blends government experience with Silicon Valley and private‑equity expertise, aiming to mitigate AI supply‑chain risk while accelerating the integration of advanced models such as Anthropic’s Claude, OpenAI’s ChatGPT, and emerging competitors.
Key members include:
- Emil Michael – former Uber COO, former White House Fellow, and current Pentagon CTO.
- Steve Feinberg – founder of Cerberus Capital Management and Deputy Secretary of Defense (Research & Engineering).
Their combined influence is already reshaping procurement, policy, and the broader AI policy landscape.
Why the Pentagon Created an AI Advisory Squad
In early 2025, the DoD issued a National Security Memorandum mandating a diversified AI supply chain. The memo required at least two cleared AI providers for classified work, a direct response to the “single‑supplier vulnerability” exposed by recent cyber‑espionage incidents.
To operationalize this directive, Secretary of Defense Pete Hegseth appointed a cross‑functional team that could:
- Evaluate frontier AI models for security, reliability, and ethical compliance.
- Negotiate contracts that balance cost, speed, and strategic independence.
- Integrate AI capabilities into existing war‑fighting platforms without compromising classified data.
These goals align with the Enterprise AI platform by UBOS, which offers a modular approach to secure AI deployment across large organizations.
The Two Power Players
Emil Michael – From Uber’s “A‑Team” to Pentagon CTO
Michael’s résumé reads like a Silicon Valley thriller: he rose to become Uber’s second‑in‑command, survived a high‑profile exit in 2017, and later served as a White House Fellow under the Obama administration. During that fellowship, he spent two years as a special assistant to Defense Secretary Robert Gates, gaining firsthand exposure to defense procurement processes.
At the Pentagon, Michael now leads the OpenAI ChatGPT integration effort, championing the use of generative AI for rapid intelligence analysis while insisting on strict data‑isolation protocols.
Critics point to his controversial past—particularly the “God Mode” tracking tool and alleged “A‑Team” culture at Uber—but supporters argue his deep understanding of large‑scale data pipelines is precisely what the DoD needs to harness AI at scale.
Steve Feinberg – Private‑Equity Titan Turned Defense Strategist
Feinberg founded Cerberus Capital Management, a $65 billion asset manager known for acquiring distressed assets and turning them into profitable enterprises. His portfolio includes defense contractors such as DynCorp and a venture arm that funds early‑stage national‑security startups.
Appointed Deputy Secretary of Defense (Research & Engineering) in 2024, Feinberg leverages his investment acumen to evaluate AI vendors not just on technology but on long‑term financial stability—a crucial factor when the Pentagon seeks to avoid “single‑supplier” lock‑ins.
Feinberg’s presence signals a shift toward market‑driven procurement, echoing the UBOS partner program model that aligns vendor incentives with government outcomes.
Strategic Implications: What This Means for U.S. Defense
The advisory team’s decisions will reverberate across three core domains:
1. AI Supply‑Chain Resilience
By insisting on at least two cleared AI providers, the Pentagon reduces the risk of a catastrophic supply‑chain breach. This approach mirrors the Chroma DB integration strategy, where redundancy is built into data storage layers.
2. Ethical and Operational Guardrails
Michael’s background in data governance informs new ChatGPT and Telegram integration policies that enforce “human‑in‑the‑loop” controls for lethal autonomous weapons, aligning with emerging international norms.
3. Accelerated Innovation Cycle
Feinberg’s venture‑capital lens encourages rapid prototyping through the UBOS templates for quick start. Start‑up‑friendly contracts allow the DoD to field AI‑enhanced decision‑support tools within months rather than years.
Collectively, these moves aim to keep the United States ahead of peer competitors while safeguarding classified data and maintaining public trust.
Key Quotes from the Verge Report
“The only reason we’re still talking to these people [Anthropic] is we need them and we need them now. The problem for these guys is that they are that good,” a senior Defense official told Axios.
“I’ve spent my career building companies that can survive market turbulence. The Pentagon can’t afford a single‑supplier failure,” Steve Feinberg said during a closed‑door briefing.
These statements underscore the urgency of diversifying AI providers while acknowledging the technical superiority of certain models—a tension that Michael and Feinberg must navigate.
Read the Full Story on The Verge
For a comprehensive account of the meeting, see the original article on The Verge: Pentagon AI bro squad includes a former Uber executive and a private‑equity billionaire.
How UBOS Supports Defense AI Initiatives
UBOS offers a suite of tools that align perfectly with the Pentagon’s new AI roadmap:
- UBOS platform overview – a low‑code environment for rapid AI model deployment.
- AI marketing agents – demonstrate how autonomous agents can manage data pipelines, a concept directly translatable to defense logistics.
- Workflow automation studio – enables secure, auditable AI workflows that meet DoD compliance standards.
- Enterprise AI platform by UBOS – provides the multi‑tenant security architecture required for classified AI workloads.
These capabilities illustrate how commercial AI platforms can be repurposed for national‑security missions without sacrificing agility.
Relevant UBOS Templates for Defense Teams
Developers can accelerate prototype building with ready‑made templates:
- AI SEO Analyzer – showcases real‑time data extraction, useful for threat‑intelligence dashboards.
- AI Video Generator – can be adapted for rapid generation of training simulations.
- AI Chatbot template – forms the basis for secure, conversational interfaces for field operators.
- AI Audio Transcription and Analysis – vital for processing intercepted communications.
Conclusion: A New Era for Defense AI
The Pentagon’s AI advisory team, anchored by Emil Michael and Steve Feinberg, is a decisive step toward a resilient, ethically grounded, and technologically superior U.S. defense posture. Their blend of government insight, Silicon Valley scale, and private‑equity rigor promises to mitigate supply‑chain risk while accelerating the fielding of next‑generation AI tools.
For organizations looking to align with this emerging ecosystem, the UBOS solutions for SMBs and the broader UBOS for startups program provide a low‑code bridge to the same standards the Pentagon is adopting.
Stay informed, adopt secure AI practices, and consider partnering with platforms that already meet defense‑grade requirements.