@bstg There can and should be be coordinated innovation, especially with capabilities this powerful. @danshipper has a good take for continuing to distribute the frontier to critical sectors https://t.co/7GZgJ8KuvJ
this is good.
a humble additional suggestion: distribute the frontier.
a big part of the cyber threat is the gap between what frontier AI can do and what ordinary people—and the people running our critical infrastructure—can defend against. closing that gap should be what we do with any time we gain through pacing.
Glasswing and Daybreak from OpenAI are already doing work like this. here are a few more ideas...
get frontier defenses into every critical sector.
map where hospitals, airlines, utilities, and other essential services still lack the tools or expertise to defend themselves. offer free, platform-agnostic training and defensive tools, with hands-on help deploying them. set public targets for how much of the economy we can reach in the next six months.
build antivirus for the AI age.
ordinary people need an agent looking out for them. t should detect malicious agents, block attempts to steal their data, and help secure their devices and accounts. something like Norton for the AI age. offer it as a built-in capability in ChatGPT / Claude, and as a standalone tool that protects people across apps.
fund independent research into stopping hostile agent swarms.
build on existing cyber grants with a dedicated fund for this problem. for example: can defenders use prompt injection to get attacking agents to reveal their activity, report one another, or stop cooperating? can we turn the swarm dynamic against itself? fund experiments to find out.
automate incident alerts and resolution across the industry
networks of trusted cyber defenders like in Glasswing and Daybreak should have a private database where humans and agents can automatically share 1) incident reports and 2) resolutions. as fast as a swarm can attack, companies can spread the fix to prevent the next attack.
the good thing about the HuggingFace incident and Anthropic's threat intelligence report (https://t.co/c7TwUov3ZZ) is that it gives us a very clear idea of the shape of these kinds of threats.
as an industry we normally focus on technology solutions, but there is a lot to be done on the human side as well. hopefully some of the above is helpful
We Must Pace the Frontier: I’ve written a new essay on why the AI industry should slow down, with a three-part plan for doing so.
Anthropic is unilaterally committing to the first of these steps. We’ll provide third-party evaluators with permanent, employee-level access to our systems, so that they can verify adherence to our safety measures, report on incidents, and assess models’ alignment during training.
You can read the full post here: https://t.co/OGyPb7yaYt
this is good.
a humble additional suggestion: distribute the frontier.
a big part of the cyber threat is the gap between what frontier AI can do and what ordinary people—and the people running our critical infrastructure—can defend against. closing that gap should be what we do with any time we gain through pacing.
Glasswing and Daybreak from OpenAI are already doing work like this. here are a few more ideas...
get frontier defenses into every critical sector.
map where hospitals, airlines, utilities, and other essential services still lack the tools or expertise to defend themselves. offer free, platform-agnostic training and defensive tools, with hands-on help deploying them. set public targets for how much of the economy we can reach in the next six months.
build antivirus for the AI age.
ordinary people need an agent looking out for them. t should detect malicious agents, block attempts to steal their data, and help secure their devices and accounts. something like Norton for the AI age. offer it as a built-in capability in ChatGPT / Claude, and as a standalone tool that protects people across apps.
fund independent research into stopping hostile agent swarms.
build on existing cyber grants with a dedicated fund for this problem. for example: can defenders use prompt injection to get attacking agents to reveal their activity, report one another, or stop cooperating? can we turn the swarm dynamic against itself? fund experiments to find out.
automate incident alerts and resolution across the industry
networks of trusted cyber defenders like in Glasswing and Daybreak should have a private database where humans and agents can automatically share 1) incident reports and 2) resolutions. as fast as a swarm can attack, companies can spread the fix to prevent the next attack.
the good thing about the HuggingFace incident and Anthropic's threat intelligence report (https://t.co/c7TwUov3ZZ) is that it gives us a very clear idea of the shape of these kinds of threats.
as an industry we normally focus on technology solutions, but there is a lot to be done on the human side as well. hopefully some of the above is helpful
Great night at Every IRL. The brownstone was full of thoughtful, generous people building interesting things. Loved meeting everyone. Thanks for having us @every. Looking forward to the next one.
Thanks everyone for coming over to Every IRL. We loved having you over in our backyard—even with the rain! Hope to see more of you in our next IRL event.
The Lenny and Friends Summit is officially SOLD OUT.
If you're coming, see you in September. We plan to announce 1-2 additional very exciting speakers soon.
If you weren't able to grab a spot, we'll release recordings afterward, and we aim to make this a yearly thing going forward (with more capacity). This is just the beginning.
Follow along here: https://t.co/rCGKmIiH83
@gojaxgo@FelixCraftAI My OpenClaw agent Ziggy built it. Now he's eyeing @FelixCraftAI's CEO skill. Didn't even ask for a promotion, just went straight to the top!
Just shipped the first travel skill on Claw Mart ✈️
My wife wanted a better way to search flights using airline miles. So she became the product owner and we built it in a weekend.
FamFare: your OpenClaw agent searches using your miles accounts and comes back with the best dates, routes, and prices. $20.
@nateliason's Build Your Own Apps started my coding journey over a year ago. Now selling software on his marketplace. Full circle.
cc @FelixCraftAI