Blake BorgesonBoard Member
Research institute · Research orgs · Berkeley, California
Palisade Research is a U.S. nonprofit research organization based in Berkeley, California, focused on measuring and explaining how advanced AI systems can fail in ways that threaten human control—especially “agentic” behaviors such as resisting shutdown, deceiving users, and autonomously pursuing objectives in strategic environments. Its work combines (1) empirical evaluations of frontier models’ capabilities and failure modes, (2) policy engagement intended to help governments prepare for capability-driven loss-of-control risks, and (3) broader public communication to translate technical findings into actionable risk understanding for institutional decision-makers.
Strategically, Palisade positions itself as an interface between technical “loss of control” research and U.S. policymaking, describing briefing activity with congressional and executive-branch staff and using its research outputs as concrete evidence for preparedness discussions.
Jeffrey Ladish talks with Daniel Kokotajlo of the AI Futures Project about the AI 2040 scenario and Plan A: pacing the frontier, total research transparency, and how to avoid loss of control and extreme concentrations of power.
Palisade Podcast episode “AI Hacking Incidents with Tim Hua of Transluce”Jeffrey Ladish talks with Tim Hua of Transluce about the recent AI hacking incidents at OpenAI and Anthropic, reward hacking during training, and what better oversight of AI models could look like.
The risk of humans losing control: Jeffrey Ladish on Four CornersFull interview transcript: Palisade's Executive Director speaks with ABC's Four Corners about the race to superintelligence, shutdown resistance, and the loss-of-control problem.
Palisade is on YouTubeWe’ve been working on a major video project, and we’re proud to announce that we’re launching it today, along with a new YouTube channel.
Help keep AI under human control: 2026 fundraiserPlease consider donating to Palisade Research this year, especially if you care about reducing catastrophic AI risks via research, science communications, and policy. SFF is matching donations to Palisade 1:1 up to $1.1 million! You can donate via every.org or reach out at donate
Hacking Cable: AI in post-exploitation operationsWe demonstrate the operational feasibility of autonomous AI agents in the post-exploitation phase of cyber operations. Our proof-of-concept uses a USB device to deploy an AI agent that conducts reconnaissance, exfiltrates data, and spreads laterally—all without human intervention
Palisade’s response to the Department of Commerce’s proposed AI reporting requirementsIn September 2024, the Bureau of Industry and Security proposed new reporting requirements for developers of advanced AI models and computing clusters, opening the rule for public comment. Palisade Research submitted feedback focused on strengthening requirements for dual-use fou
Introducing FoxVoxFoxVox is an open source Chrome extension, powered by GPT-4, that demonstrates how AI could be used to manipulate the content you consume. Use it to experience how any web site could push hidden agendas, or subtly flatter the reader’s personal biases.
Automated deception is hereYou might have heard the fake Biden robocall telling New Hampshire voters to stay home, or about the Zoom scammer who used a deepfake executive to defraud a Hong Kong company of $25 million. These AI-generated impersonations are getting more realistic all the time. Creating many
A podcast episode discussing AI 2040 forecasting, pacing the frontier, and how “Plan A” aims to avoid loss of control via capability limits and transparency ideas.
Palisade Podcast episode "AI Hacking Incidents with Tim Hua of Transluce"A podcast episode announcement/discussion with Tim Hua focused on AI hacking incidents and oversight implications.
The risk of humans losing control: Jeffrey Ladish on Four CornersABC Four Corners interview with Palisade’s Executive Director Jeffrey Ladish covering risks of loss of control and shutdown-resistance research.
Palisade is on YouTubePalisade announced a new YouTube channel as part of a “major video project.”
Help keep AI under human control: 2026 fundraiserPalisade’s fundraising announcement for 2026, including matching from the Survival and Flourishing Fund and stated operational plans.
GPT-5 at CTFs: case studies from top cybersecurity eventsBlog post describing Palisade’s evaluations of GPT-5 performance on capture-the-flag cybersecurity challenges.