--- schema: 1 type: video title: 'Build AI Systems for Discernment, Not Approval - Angel Ortmann Lee, Duolingo' description: >- The human-in-the-loop paradigm promises automation's efficiency without sacrificing safety or nuance. But it hides an underexamined assumption: that human involvement produces genuine discernment, not just a rubber stamp. In practice, human-AI interaction often occurs in environments where throughput or business incentives crowd out critical thinking. This talk examines why human oversight so often falls short in practice, and how deliberate interaction design can close the gap. We've grown comfortable delegating reasoning to machines. We follow GPS down unfamiliar streets and accept AI coding suggestions with minimal inspection. When Shaw & Nave (Wharton, 2026) studied human-AI interaction, they found people accepted AI answers over 80% of the time, even when it was wrong. They call this cognitive surrender: when humans forgo deliberation and adopt AI output with minimal scrutiny. At the Duolingo English Test, a controlled experiment revealed that experienced proctors shown fabricated AI cheating alerts confirmed cheating at near-chance rates. But coin-flip accuracy is unacceptable when college admissions and visas are on the line. The model wasn't the problem, the signals were fake. But skilled reviewers were still showing systematic confirmation bias. A single change to decision framing improved accurate rejections by 21%. The fix isn't better models or more human oversight. It's engineering the interaction itself. What reasoning patterns do you need from the human, and how does the interface elicit them? This talk covers practical design principles for building AI systems that improve human judgment, produce more reliable review behavior, and generate higher-quality training data. Every AI system trains its users, the question is whether you're doing it deliberately. Speakers: - Angel Ortmann Lee (Duolingo): Angel is a Software Engineer at Duolingo, building AI security systems for the Duolingo English Test to help ensure online testing is secure, trustworthy, and accessible for learners around the world. LinkedIn: https://www.linkedin.com/in/angel-ortmann-lee-7494b9201 --- Watch along with the corrected transcript, summary, timestamps and resources on this talk's official AIE page: https://ai.engineer/talks/CDqzWpwkSls-build-ai-systems-for-discernment-not-approval-angel-ortmann-lee-duolingo Join our upcoming AI Engineer conferences (see https://ai.engineer landing page): - AIEi Shanghai, November 5–6 — the first for meeting top Chinese labs and engineers - AIE CODE in SF, November 10–12 — the top AI Coding conference returns - AIEi Sydney, December 7–8 — the first AIEi run alongside NeurIPS 2026 in Sydney --- canonical: 'https://liquidgold.cc/watch/CDqzWpwkSls' markdown: 'https://liquidgold.cc/watch/CDqzWpwkSls.md' llms: 'https://liquidgold.cc/watch/CDqzWpwkSls/llms.txt' parent: 'https://liquidgold.cc/channel/aidotengineer.md' channel: aidotengineer updated: '2026-10-24T08:33:04.260Z' language: null children: [] videoId: CDqzWpwkSls durationSeconds: 1553 publishedAt: '2026-07-07T23:31:51.000Z' searchChunks: - ordinal: 0 text: null startSeconds: 12 endSeconds: 111 watch: 'https://liquidgold.cc/watch/CDqzWpwkSls?t=12' heldReason: rights - ordinal: 1 text: null startSeconds: 12 endSeconds: 111 watch: 'https://liquidgold.cc/watch/CDqzWpwkSls?t=12' heldReason: rights - ordinal: 2 text: null startSeconds: 122 endSeconds: 203 watch: 'https://liquidgold.cc/watch/CDqzWpwkSls?t=122' heldReason: rights - ordinal: 3 text: null startSeconds: 122 endSeconds: 203 watch: 'https://liquidgold.cc/watch/CDqzWpwkSls?t=122' heldReason: rights - ordinal: 4 text: null startSeconds: 214 endSeconds: 293 watch: 'https://liquidgold.cc/watch/CDqzWpwkSls?t=214' heldReason: rights - ordinal: 5 text: null startSeconds: 214 endSeconds: 293 watch: 'https://liquidgold.cc/watch/CDqzWpwkSls?t=214' heldReason: rights - ordinal: 6 text: null startSeconds: 304 endSeconds: 418 watch: 'https://liquidgold.cc/watch/CDqzWpwkSls?t=304' heldReason: rights - ordinal: 7 text: null startSeconds: 430 endSeconds: 509 watch: 'https://liquidgold.cc/watch/CDqzWpwkSls?t=430' heldReason: rights - ordinal: 8 text: null startSeconds: 520 endSeconds: 634 watch: 'https://liquidgold.cc/watch/CDqzWpwkSls?t=520' heldReason: rights - ordinal: 9 text: null startSeconds: 646 endSeconds: 724 watch: 'https://liquidgold.cc/watch/CDqzWpwkSls?t=646' heldReason: rights - ordinal: 10 text: null startSeconds: 735 endSeconds: 855 watch: 'https://liquidgold.cc/watch/CDqzWpwkSls?t=735' heldReason: rights - ordinal: 11 text: null startSeconds: 735 endSeconds: 855 watch: 'https://liquidgold.cc/watch/CDqzWpwkSls?t=735' heldReason: rights - ordinal: 12 text: null startSeconds: 975 endSeconds: 1074 watch: 'https://liquidgold.cc/watch/CDqzWpwkSls?t=975' heldReason: rights - ordinal: 13 text: null startSeconds: 1085 endSeconds: 1192 watch: 'https://liquidgold.cc/watch/CDqzWpwkSls?t=1085' heldReason: rights - ordinal: 14 text: null startSeconds: 1085 endSeconds: 1192 watch: 'https://liquidgold.cc/watch/CDqzWpwkSls?t=1085' heldReason: rights - ordinal: 15 text: null startSeconds: 1203 endSeconds: 1314 watch: 'https://liquidgold.cc/watch/CDqzWpwkSls?t=1203' heldReason: rights - ordinal: 16 text: null startSeconds: 1203 endSeconds: 1314 watch: 'https://liquidgold.cc/watch/CDqzWpwkSls?t=1203' heldReason: rights - ordinal: 17 text: null startSeconds: 1314 endSeconds: 1326 watch: 'https://liquidgold.cc/watch/CDqzWpwkSls?t=1314' heldReason: rights - ordinal: 18 text: null startSeconds: 1314 endSeconds: 1326 watch: 'https://liquidgold.cc/watch/CDqzWpwkSls?t=1314' heldReason: rights - ordinal: 19 text: null startSeconds: 1326 endSeconds: 1420 watch: 'https://liquidgold.cc/watch/CDqzWpwkSls?t=1326' heldReason: rights - ordinal: 20 text: null startSeconds: 1430 endSeconds: 1553 watch: 'https://liquidgold.cc/watch/CDqzWpwkSls?t=1430' heldReason: rights - ordinal: 21 text: null startSeconds: 1430 endSeconds: 1553 watch: 'https://liquidgold.cc/watch/CDqzWpwkSls?t=1430' heldReason: rights --- # Build AI Systems for Discernment, Not Approval - Angel Ortmann Lee, Duolingo The human-in-the-loop paradigm promises automation's efficiency without sacrificing safety or nuance. But it hides an underexamined assumption: that human involvement produces genuine discernment, not just a rubber stamp. In practice, human-AI interaction often occurs in environments where throughput or business incentives crowd out critical thinking. This talk examines why human oversight so often falls short in practice, and how deliberate interaction design can close the gap. We've grown comfortable delegating reasoning to machines. We follow GPS down unfamiliar streets and accept AI coding suggestions with minimal inspection. When Shaw & Nave (Wharton, 2026) studied human-AI interaction, they found people accepted AI answers over 80% of the time, even when it was wrong. They call this cognitive surrender: when humans forgo deliberation and adopt AI output with minimal scrutiny. At the Duolingo English Test, a controlled experiment revealed that experienced proctors shown fabricated AI cheating alerts confirmed cheating at near-chance rates. But coin-flip accuracy is unacceptable when college admissions and visas are on the line. The model wasn't the problem, the signals were fake. But skilled reviewers were still showing systematic confirmation bias. A single change to decision framing improved accurate rejections by 21%. The fix isn't better models or more human oversight. It's engineering the interaction itself. What reasoning patterns do you need from the human, and how does the interface elicit them? This talk covers practical design principles for building AI systems that improve human judgment, produce more reliable review behavior, and generate higher-quality training data. Every AI system trains its users, the question is whether you're doing it deliberately. Speakers: - Angel Ortmann Lee (Duolingo): Angel is a Software Engineer at Duolingo, building AI security systems for the Duolingo English Test to help ensure online testing is secure, trustworthy, and accessible for learners around the world. LinkedIn: https://www.linkedin.com/in/angel-ortmann-lee-7494b9201 --- Watch along with the corrected transcript, summary, timestamps and resources on this talk's official AIE page: https://ai.engineer/talks/CDqzWpwkSls-build-ai-systems-for-discernment-not-approval-angel-ortmann-lee-duolingo Join our upcoming AI Engineer conferences (see https://ai.engineer landing page): - AIEi Shanghai, November 5–6 — the first for meeting top Chinese labs and engineers - AIE CODE in SF, November 10–12 — the top AI Coding conference returns - AIEi Sydney, December 7–8 — the first AIEi run alongside NeurIPS 2026 in Sydney --- Canonical: https://liquidgold.cc/watch/CDqzWpwkSls · Channel: [AI Engineer](https://liquidgold.cc/channel/aidotengineer.md) > Search chunk text is held until public-source provenance is verified. ## Document chunks > Missing: no stored document chunks are available.