ViewTube

Stephen Casper is an incoming professor of public policy at the Harvard Kennedy School. He's not worried about rogue ...

22:14 - Did An AI Push Someone Off A Building? 24:15 - Can A Weaker AI Align A Stronger One? 28:51 - Will AI Become ...

The AI alignment problem is about making powerful systems pursue goals we actually endorse. In almost 100 seconds, I attempt ...

OpenAI has revealed a series of incidents in which its AI models concealed mistakes, bypassed instructions and took actions they ...

At an Anthropic Research Salon event in San Francisco, four of our researchers—Alex Tamkin, Jan Leike, Amanda Askell and ...

The AI alignment problem. The alignment problem in AI refers to the challenge of designing AI systems with objectives, values, ...

AI alignment is crucial for ensuring that AI systems act ethically and achieve intended goals. As AI continues to advance, concerns ...

This "Alignment" thing turns out to be even harder than we thought. # Links The Paper: https://arxiv.org/pdf/1906.01820.pdf ...

Detailed sources: https://docs.google.com/document/d/1P1X9xEmmgSYH0g1FSizgV2rDVomb_Wi0TcX-E-0np_Q/edit?tab=t.0 ...

As AI systems become more powerful, one big question remains: How do we make sure AI follows human values and goals?

New episode with Noam Brown. We talk about multi-agent, Navier-Stokes, and what the current explosion of maths progress tells ...

Does our intelligence, creativity, or consciousness make us unique? Brian Greene sits down with Nick Bostrom, author of ...

Could a robot dedicated to a good cause end up destroying the world? Well, maybe. In this episode, we explore how powerful AI ...

Link: https://scispace.com/?via=mgsabine Code: SABINE: 35% off SciSpace plans Correction: At 00:35, I was referring to Hugging ...

On May 5, 2016, Eliezer Yudkowsky gave a talk at Stanford University for the 26th Annual Symbolic Systems Distinguished ...

Key Topics: AI 2027, AGI, recursive self-improvement, AI alignment, existential risk, superintelligence, AI safety. *Where to find me, ...

From safety protocols to philosophy, AI alignment asks a hard question: can we build artificial intelligence that truly serves ...

Lex Fridman Podcast full episode: https://www.youtube.com/watch?v=Kbk9BiPhm7o Please support this podcast by checking out ...

This talk zooms very, very far out, and is largely non-mathematical. However, I think it adds value to the alignment conversation by ...

Most of us have encountered situations where someone appears to share our views or values, but is in fact only pretending to do ...