514 karma · joined March 9, 2019
The artificial intelligence startup Anthropic, founded by former OpenAI employees concerned about A.I. safety, is releasing a new chatbot called Claude. However, the company's employees remain deeply worried about the risks of powerful A.I. models like Claude becoming dangerous if misused. They compare themselves to the creators of the atomic bomb and obsess over existential threats that advanced A.I. could pose. Still, they argue that building cutting-edge models is necessary for researching how to make them safer, and that having safety-focused companies in the A.I. race is better than leaving it solely to profit-driven firms. So Anthropic pushes forward, attempting to balance competitiveness with caution.
Should we be using if an email is pwned as input to antiabuse systems to give them higher confidence?
It reminds me a bit of when the % of emails that were #spam vs ham crossed 50% many years ago.
I have been using poised (https://www.poised.com/) for this as it does it and related things...but have backed off on it since they went from free to paid (though is still may pay for it).
There is no question that next-generation ‘auto-complete’ tools like GitHub Copilot will increase the productivity of software developers. However, while Copilot can rapidly generate prodigious amounts of code, our conclusions reveal that developers should remain vigilant (‘awake’) when using Copilot as a co-pilot. Ideally, Copilot should be paired with appropriate security-aware tooling during both training and generation to minimize the risk of introducing security vulnerabilities. While our study provides new insights into its behavior in response to security-relevant scenarios, future work should investigate other aspects, including adversarial approaches for security-enhanced training
Launching attacks during major news events surely also helped the attackers stay under the radar for longer.
https://www.wired.com/story/surviving-as-an-old-in-the-tech-...
https://www.linkedin.com/pulse/google-leaders-greyglers-navi...
https://www.amazon.com/Wisdom-Work-Making-Modern-Elder/dp/05...
These are from https://about.gitlab.com/company/culture/inclusion/tmrg-gitl...