What responsibilities should AI creators and firms have for the behavior and impact of their models, and why?
Most answers emphasize accountability, legal compliance, and risk mitigation for AI creators, though some highlight transparency, fairness, and the need for external oversight or self-regulation.
12 answers
Answers


Dean Ball
Head of Strategic Futures · Openai
Accountability for Misuse
Dean Ball says AI developers should be accountable for unsafe development practices and that people and firms using AI should be held accountable when their misuse or lack of oversight causes harm, arguing this requires legal and technical infrastructure tying people to their agents' actions.
Don't Worry About the Vase · Don't Worry About the Vase · Read

Ali Partovi
CEO · Neo
Hold them as if intentional
Ali Partovi says the law should hold AI companies responsible when their AI does something illegal and treat it as if they did it intentionally, at the highest penalty, because treating such incidents as accidents keeps them happening.
MTS · Listen on Apple Podcasts

Reid Hoffman
Co-Founder, Executive Board Chair · Manas Ai
Ensure accountability
Reid Hoffman says figuring out accountability and accountability laws is important, and that the right place to start is to commission smart work on a dynamic framework for evolving accountability.
Reid Hoffman · Watch on YouTube

Tyler Cowen
Professor · George Mason University
Make models publicly available
Tyler Cowen says that if companies use powerful models internally for research or engineering, they should also make them publicly available, because he thinks good decisions can't be made without access to these models.
Interesting Times with Ross Douthat · Watch on YouTube
Daron Acemoglu
Institute Professor · Massachusetts Institute of Technology
Ensure fairness and compensate data contributors
Daron Acemoglu says firms should compensate the workers who generate their data, calling that fairness and also a good business decision.
The answer starts at 41:46.
Silicon Valley Girl: AI, Tech and Career Growth · Listen on Apple Podcasts

Tom Bilyeu
CEO · Impact Theory
Ensure Safety
Tom Bilyeu says AI company leaders like Dario have a duty to employees and shareholders to make the company successful but also a responsibility to do things safely.
Tom Bilyeu's Impact Theory · Tom Bilyeu's Impact Theory · Listen at 6:22

Vlad Tenev
Co-Founder, Chairman, CEO · Robinhood
Be Held Responsible
Vlad Tenev does not give this answer in the passage; the quoted remarks about liability and being held responsible come from Treasury Secretary Scott Bessent, not Tenev.
The answer starts at 4:25.
Moonshots with Peter Diamandis · Listen on Apple Podcasts

Dario Amodei
CEO and Co-Founder · Anthropic
Add external evaluators and coordinate limits
Dario Amodei proposes that AI firms embed external evaluators free to publish their findings uncensored and that democratic countries coordinate to set common limits, as steps to slow the pace of AI development.
Quality Value · Quality Value · Read
Richard Socher
CEO · Recursive
Ensure legal compliance
Richard Socher says AI must ultimately be aligned with the law wherever it is deployed, framing legal alignment as the baseline requirement for AI creators.
Latent Space · Latent Space · Read

Greg Brockman
President, Chairman, & Co-Founder · Openai
Mitigate risks
Greg Brockman says AI creators must take the impacts and risks seriously and really mitigate them, while also communicating the nuance so people don't throw out the baby with the bathwater.
The answer starts at 37:39.
Odd Lots · Listen on Apple Podcasts
Anders Hejlsberg
Technical Fellow · Microsoft
Take responsibility
Anders Hejlsberg says someone has to take responsibility for what AI is doing, and if you give it the keys, it's on you, because users will sue you rather than the AI.
Ryan Peterman · Watch on YouTube

Rhett Alden
CTO, Elsevier Health Markets · Elsevier
Self‑regulate responsibly
Rhett Alden says his company, Elsevier, is establishing its own responsible AI principles and trust factors rather than waiting for regulators, because regulation can't keep up with AI's pace and self-regulation is currently the norm.
The answer starts at 5:32.
The AI in Business Podcast · Listen on Apple Podcasts
More Generative AI questions
View all →- What design choice improves AI agent privacy?5 answers
- What architectural choices enable reliable agentic AI systems?8 answers
- How does your team create feedback loops for AI models?12 answers
- What architecture optimizes compute for AI workloads?13 answers
- How do you balance AI customization and scalability?8 answers



















