Yeah, I'm one of them, but I remain deeply concerned that we are going to live to regret it!
ETHICS
-
Local AI Models Enable Sensitive Journalism Without Cloud Dependencies
By
–
One of the reasons I've been following advances in local models with so much interest – Qwen 3.5 on a well specced laptop is now powerful enough that it might be worth using for truly sensitive journalism projects
-
Anonymous Sources at Risk: Hosted AI Models and Journalism Security
By
–
Journalism is one of the few fields where I think concern about hosted models is very reasonable – if you're working with data that could expose one of your anonymous sources you have a genuine risk that a government might use a subpoena to access logged data from a provider
-
Possibility Space Exceeds Conceivable Action in AI Systems
By
–
The space of things you can do is much greater than the space of things you can conceive of doing.
-
AI Progress Measured by Intelligence Sharpe Ratio
By
–
A good measure of progress in AI is the intelligence Sharpe ratio, a system's average intelligence across tasks divided by its standard deviation. By this measure, current AIs are still very weak because their variance is so high.
-
Humility in AI: Avoiding Veneration Amid Rapid Change
By
–
Ehh I don't think that's the attitude to bring. We're all just one person each working with very new workflows, with the models changing underneath us. I'm sure he would tell you not to venerate.
-
Agentic AI and Intelligence Explosion Essay
By
–
"The recent explosion of agentic AI suggests the possibility of something similar at the scale of billions of interacting minds, human and non-human alike."
An essay on the Intelligence Explosion @ScienceMagazine by @profjamesevans @bratton @blaiseaguera -
Anthropic’s System Cards Reveal Remaining AI Security Vulnerabilities
By
–
Anthropic's own system cards still show attacks making it through
-
AI Models Trained to Resist Prompt Injection Attacks
By
–
I think both – plus the labs have been putting a lot of effort into training them to resist prompt injection style attacks Anthropic usually mention prompt injection in their system cards eg this one for Opus 4.6 https://
www-cdn.anthropic.com/0dd865075ad313
2672ee0ab40b05a53f14cf5288.pdf
… -
Three AI Eras: From Concentration to Recursive Self-Improvement
By
–
When I think about AI power, I see 3 eras: Era 1 (now): Only a few companies can build frontier AI. They hold the cards.
Era 2: Recursive self-improving AI arrives – AI that builds better AI on its own. Short dangerous window. The leaders will try to grab real-world