You're putting words in my mouth, and totally misunderstanding the point. I did not say I think we never get there. I simply said that experts on nuclear safety should be considered when discussing nuclear safety.
@jeremyphoward
-
Nuclear Experts Should Be Heard in AI Risk Discussions
By
–
I agree – I certainly don't want him to stop speaking. But I do think nuclear experts should also be listened to when considering nuclear analogies. There are many other experts, in fact, with relevant expertise.
-
No Great Precedents Exist for AI’s Transformative Impact
By
–
There aren't any great ones — in the sense there isn't some existing thing we could borrow and copy over all our understanding and experience and expect it to "just work". Somewhat similar things include: nanotech; crispr; LHC; recombinant DNA; cryptography.
-
Managing Unprecedented Power Growth in AI Systems
By
–
I also think we should understand that, overall, the concern is that there will be a very large increase in power in the world, and how we manage that. People that have expertise in the use of misuse of power are critical to learn from here.
-
AI risks compared to DNA editing, nanotech, and LHC fears
By
–
There were fears of total wipeout of humanity around DNA editing, nanotech, and the LHC too. AI is not unique in having such large potential impacts. I think it's helpful to draw on the expertise of those that have navigated these before us.
-
Google Whistleblowers Warn About AI Risks Years Before Public
By
–
I'd also like to see more awareness of all the people that have spoken up about risks for years, some of whom got fired from Google for doing so.
-
Managing Risks in Advanced Technology Transitions
By
–
"Risk management of things that are totally novel" is indeed complex. Thankfully, people have been studying that risk management for years. For instance, here's a call from Professor of Advanced Tech Transitions @2020science
, to learn from this research: https://
andrewmaynard.substack.com/p/navigating-a
dvanced-technology-transitions
… -
Collaboration and Diverse Skills Key to Effective AI Management
By
–
Managing this effectively will require a large set of diverse skills and experiences that no one person possesses, so the key is collaboration and openness.
-
Corporate AI Overreach Requires Thoughtful Regulation
By
–
It's important to understand that the push back against the current corporate over-reach is largely from people who *support* thoughtful regulation of high risk applications and harmful algorithmic decision making. It's not an anti-regulation stance.
-
EU AI Act Shifted from Deployment to R&D Regulation
By
–
In fact the EU AI Act up until the very last minute *did* focus entirely on regulating product deployment. But then all the lobbying happened and the act was overhauled to regulate R&D. Back when it focused on high risk applications, instead of R&D, it looked great!