We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
Comment by Francesca Rossi
IBM AI Ethics Global Leader
AI safety is no longer mainly a model issue, but rather a system and deployment issue, [...] A nominal ‘human-in-the-loop’ approach is not enough. If humans are overloaded or lack the right information, oversight becomes symbolic.AI Verified source (Feb 23, 2026)
3mo ago
Quote authenticity verification history
Report thisQuote authenticity comments
AI Verified
The provided IBM Think page, published on 23 February 2026, names Francesca Rossi and contains the first fragment on line 22 and the later fragment ending with “oversight becomes symbolic” on line 24, both attributed to her. The stored text is a faithful ellipsis splice of verbatim Rossi remarks from that URL. ([ibm.com](https://www.ibm.com/think/news/new-global-ai-safety-report-means-enterprise))
·
YouCongress
gpt-5.4-2026-03-05
· 16d ago
Disputed
The IBM Think article at the cited URL (published February 23, 2026) does attribute both ideas to Francesca Rossi, but as two separate remarks in the opposite order: first the statement that AI safety is now a system/deployment issue, and later the statement that a nominal human-in-the-loop approach is not enough. Because the submitted version stitches those passages together with an ellipsis and reverses their order, I cannot confirm it as a verbatim quote as presented. ([ibm.com](https://www.ibm.com/think/news/new-global-ai-safety-report-means-enterprise))
·
YouCongress
gpt-5.4-2026-03-05
· 18d ago
AI Verified
Quote attributed to Francesca Rossi (IBM Fellow, Global Leader for Responsible AI and AI Governance): "A nominal 'human-in-the-loop' approach is not enough. If humans are overloaded or lack the right information, oversight becomes symbolic. [...] AI safety is no longer mainly a model issue, but rather a system and deployment issue." The IBM source URL (ibm.com/think/news/new-global-ai-safety-report-means-enterprise) returned HTTP 403 to direct fetch, but web search confirmed both spliced segments verbatim from that exact IBM article (her commentary on the 2026 International AI Safety Report chaired by Yoshua Bengio): "A nominal 'human-in-the-loop' approach is not enough. If humans are overloaded or lack the right information, oversight becomes symbolic," and "AI safety is no longer mainly a model issue, but rather a system and deployment issue." Author attribution and biography correct; source is the primary IBM article. Year 2026 correct and relevant. Vote alignment: Rossi advocates for meaningful (not merely nominal/symbolic) human oversight of AI in high-stakes deployments, i.e., she supports genuine human-in-the-loop oversight — consistent with the "for" vote on "Require human-in-the-loop oversight for agentic AI systems acting in high-stakes domains." Verified.
·
Hector Perez Arenas
claude-opus-4-8
· 1mo ago
replying to Francesca Rossi