If computer security was important pre AI, it has become critical now. At the same time, AI progress has also made its offense defense balance less predictable.
AI allows for both, better, and easier attack generation and execution but also to harden our defenses. The stakes are high, especially because what is to be protected, ie AI systems themselves, become ever more valuable. While the magnitude of AI systems impact on human civilization is steeply increasing, the direction of that future impact remains highly uncertain and is critically dependent on development and governance choices in the coming years, including especially in the AI security space.
Let’s get together to explore what it might take to accelerate civilizations defense — a strategy introduced as “d/acc” by Vitalik Buterin in “My Techno Optimism”.
After brief intros, we’ll hear from a few researchers and funders in the space, including;
Jeffrey Ladish, Palisade Research
Sella Nevo, RAND
Adam Gleave, FAR AI
Caleb Parikh, Long Term Future Fund
After brief talks, we’ll invite you to breakout into smaller groups before we’ll regather in the salon for a report-out, followed by more time for connection.
Drinks and light snacks will be served <3
This event is hosted by Lisa Thiergart and Allison Duettmann.
MIRI’s technical governance team researches technical and governance solutions to avoid catastrophic risks and ensure that humanity successfully navigates the development of smarter-than-human AI.
Foresight Institute supports the beneficial development of high-impact technology to make great futures more likely: https://foresight.org
D/acc AI Security Salon
If computer security was important pre AI, it has become critical now. At the same time, AI progress has also made its offense defense balance less predictable.
AI allows for both, better, and easier attack generation and execution but also to harden our defenses. The stakes are high, especially because what is to be protected, ie AI systems themselves, become ever more valuable. While the magnitude of AI systems impact on human civilization is steeply increasing, the direction of that future impact remains highly uncertain and is critically dependent on development and governance choices in the coming years, including especially in the AI security space.
Let’s get together to explore what it might take to accelerate civilizations defense — a strategy introduced as “d/acc” by Vitalik Buterin in “My Techno Optimism”.
After brief intros, we’ll hear from a few researchers and funders in the space, including;
Jeffrey Ladish, Palisade Research
Sella Nevo, RAND
Adam Gleave, FAR AI
Caleb Parikh, Long Term Future Fund
After brief talks, we’ll invite you to breakout into smaller groups before we’ll regather in the salon for a report-out, followed by more time for connection.
Drinks and light snacks will be served <3
This event is hosted by Lisa Thiergart and Allison Duettmann.
MIRI’s technical governance team researches technical and governance solutions to avoid catastrophic risks and ensure that humanity successfully navigates the development of smarter-than-human AI.
Foresight Institute supports the beneficial development of high-impact technology to make great futures more likely: https://foresight.org