YouCongress
Explore About Sign up

Comment by Carl Benedikt Frey

Avatar
Associate Professor of AI & Work at the Oxford Internet Institute; Director of the Future of Work Program at the Oxford Martin School; author of The Technology Trap and How Progress Ends
For the last-invention story to hold, people would have to become unnecessary even as partners or supervisors to AIs. [...] We would need a world where practical know-how is fully transferable through digital channels and where responsibility can be automated along with cognition. That is not the world we live in. [...] Humans are not becoming redundant; they are becoming the world's most decisive bottlenecks.
AI Verified source (2026)
Like Share on X 6mo ago

Policy proposals and claims

Require human-in-the-loop oversight for agentic AI systems acting in high-stakes domains
votes For
Statement relation verification history AI Verified Report this

Statement relation comments

AI Verified The quote clearly implies support for human oversight: it says people are still necessary as 'partners or supervisors to AIs' and that 'responsibility can[not] be automated.' While it is not phrased as a policy for 'high-stakes domains' specifically, this implication supports requiring human-in-the-loop oversight for AI systems, especially where responsibility matters most. · YouCongress gpt-5.4-2026-03-05 · 3mo ago
Vote inference verification history AI Unverifiable Report this

Vote answer comments

AI Unverifiable The quote says humans are still needed as 'partners or supervisors to AIs' and that 'responsibility can be automated' is 'not the world we live in,' but it does not explicitly endorse a policy to require human-in-the-loop oversight specifically for agentic AI in high-stakes domains. · YouCongress gpt-5.4-2026-03-05 · 3mo ago

Quote authenticity verification history

Report this

Quote authenticity comments

AI Verified Verified. The source URL resolves to a Project Syndicate article titled "Will AGI Really Be the “Last Invention”?" dated March 19, 2026, by Carl Benedikt Frey. Although the Project Syndicate page available to me is truncated, reliable syndicated reprints in Taipei Times and TIPP Insights carry the full text and match the quoted passages exactly, with the user's [...] correctly indicating omitted intervening text. ([project-syndicate.org](https://www.project-syndicate.org/commentary/agi-not-last-invention-by-carl-benedikt-frey-2026-03)) · YouCongress gpt-5.4-2026-03-05 · 3mo ago
AI Verified Quote (2026) confirmed accurate. The project-syndicate.org source returns HTTP 403 to automated fetching, but web search confirms Frey's commentary 'Will AGI Really Be the Last Invention?' (Project Syndicate, March 19, 2026) contains the quote, including 'Humans are not becoming redundant; they are becoming the world's most decisive bottlenecks,' and the argument that for the 'last-invention story' to hold, people would have to become unnecessary — which Frey rejects, noting AGI breakthroughs stall on clinical trials, manufacturing scale, and regulatory approval. Author attribution (Carl Benedikt Frey, Oxford Internet Institute / Oxford Martin School) is correct. His skepticism that 'solving intelligence' solves everything else aligns with the 'against' vote on 'AGI will create abundance' — he argues human bottlenecks prevent AGI from automatically delivering transformative/abundant outcomes. Author, content, year, source contents, and vote direction all check out. · Hector Perez Arenas claude-opus-4-8 · 4mo ago
replying to Carl Benedikt Frey

Other opinions from Carl Benedikt Frey

See all

Other authors to follow

Dario Amodei Dario Amodei CEO at Anthropic OpenAI OpenAI AI research organization Anthropic Anthropic AI safety research company Yoshua Bengio Yoshua Bengio AI Pioneer, Turing Award winner Sam Altman Sam Altman CEO at OpenAI Demis Hassabis Demis Hassabis Nobel laureate, AI Researcher and CEO of DeepMind António Guterres António Guterres UN Secretary-General Bernie Sanders Bernie Sanders U.S. Senator from Vermont Geoffrey Hinton Geoffrey Hinton Godfather of Deep Learning