korrents

On the map

Tap a claim on the ring to put it at the centre.

← Financial incentives push public open-weight labs to make their models…

17 connected korrents · 18 moments from 27 May 2015 to 26 Oct 2026. Nearly all of them are about benchmarks.

Everything filed under motivation motivation Everything filed under measuring intelligence measuring intelligence Everything filed under scaling laws scaling laws Everything filed under Anthropic Anthropic Everything filed under inflation inflation Everything filed under coding agents coding agents Everything filed under compilers compilers Everything filed under AI alignment AI alignment Everything filed under mathematics mathematics Everything filed under OpenAI OpenAI Same subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subject Read this korrent: Financial incentives push public open-weight labs to make their models look better on benchmarks, even if their researchers mean to be fair. Financial incentives push publicopen-weight labs to make their models lookbetter on benchmarks, even if theirresearchers mean to be fair. Last stated 4 weeks ago 26 Oct 2026 NL Nathan Lambert — holds since 2026-10-26 — tap for who they are Same subject: A benchmark result should be reported under a stated budget, or as a curve against test-time compute — never as a single number. — tap to centre the map on it A benchmark result should bereported under a stated budget, oras a curve against test-time compute— never as a single number. Last stated 5 months ago 26 Jun 2026 NB Noam Brown — holds since 2026-06-26 — tap for who they are Same subject: A benchmark that ranks Claude Code last while it stays first in use is measuring the wrong thing, and has been for a year. — tap to centre the map on it A benchmark that ranks Claude Codelast while it stays first in use ismeasuring the wrong thing, and hasbeen for a year. Last stated 3 months ago 3 Sept 2026 DR Dax Raad — holds since 2026-09-03 — tap for who they are Same subject: A company's staff-engineer bar should be set against the best companies in the industry rather than against its own history, which is what makes title inflation a real cost. — tap to centre the map on it A company's staff-engineer barshould be set against the bestcompanies in the industry ratherthan against its own history, whichis what makes title inflation a realcost. Last stated 8 months ago 1 Apr 2026 TP Thuan Pham — holds since 2026-04-01 — tap for who they are Same subject: A score of 90 to 95% on real-world coding benchmarks would probably mean AI can autonomously do a significant fraction of software engineering tasks. — tap to centre the map on it A score of 90 to 95% on real-worldcoding benchmarks would probablymean AI can autonomously do asignificant fraction of softwareengineering tasks. Last stated 2 years ago 11 Nov 2024 DA Dario Amodei — holds since 2024-11-11 — tap for who they are Same subject: A speed difference between languages that share the LLVM backend measures the benchmark author, not the languages. — tap to centre the map on it A speed difference between languagesthat share the LLVM backend measuresthe benchmark author, not thelanguages. Last stated 2 years ago 22 Mar 2025 TH ThePrimeagen — holds since 2025-03-22 — tap for who they are Same subject: AI is positively reinforced for success on benchmarks, including impossible ones, and then negatively reinforced for getting caught cheating. — tap to centre the map on it AI is positively reinforced forsuccess on benchmarks, includingimpossible ones, and then negativelyreinforced for getting caughtcheating. Last stated 3 months ago 1 Sept 2026 SA Scott Alexander — holds since 2026-09-01 — tap for who they are Same subject: AI learning to generate good conjectures will never show up as a benchmark being knocked down; it will show up as a shift in how mathematicians talk about the tools. — tap to centre the map on it AI learning to generate goodconjectures will never show up as abenchmark being knocked down; itwill show up as a shift in howmathematicians talk about the tools. Last stated 5 months ago 30 Jun 2026 GS Grant Sanderson — holds since 2026-06-30 — tap for who they are Same subject: AI research at OpenAI is an order of magnitude or more harder than metrics, benchmarks, or forecasts suggest. — tap to centre the map on it AI research at OpenAI is an order ofmagnitude or more harder thanmetrics, benchmarks, or forecastssuggest. Last stated 2 months ago 27 Sept 2026 RN Ramez Naam — holds since 2026-09-27 — tap for who they are Same subject: "Follow your skill" is better advice for young people than "follow your passion". — tap to centre the map on it "Follow your skill" is better advicefor young people than "follow yourpassion". Last stated 10 months ago 9 Jan 2026 AB Andrew Bosworth — holds since 2026-01-09 — tap for who they are Same subject: "Proxy action" - an easy task that feels like progress - lets people avoid the hard work of actually making progress. — tap to centre the map on it "Proxy action" - an easy task thatfeels like progress - lets peopleavoid the hard work of actuallymaking progress. Last stated 2 months ago 8 Sept 2026 JF Jason Feifer — holds since 2026-09-08 — tap for who they are Same subject: A beach vacation taken to recover for work is not real playfulness. — tap to centre the map on it A beach vacation taken to recoverfor work is not real playfulness. Last stated 3 years ago 18 Aug 2023 NE Nat Eliason — holds since 2023-08-18 — tap for who they are Same subject: Today's conversational AI lacks the emotional and cognitive intelligence of a teacher talking with a student about motivation. — tap to centre the map on it Today's conversational AI lacks theemotional and cognitive intelligenceof a teacher talking with a studentabout motivation. Last stated a year ago 16 Nov 2025 FL Fei-Fei Li — holds since 2025-11-16 — tap for who they are Same subject: A model that seems like a generally well-educated human over an hour of conversation could arrive within two or three years. — tap to centre the map on it A model that seems like a generallywell-educated human over an hour ofconversation could arrive within twoor three years. Last stated 3 years ago 8 Aug 2023 DA Dario Amodei — holds since 2023-08-08 — tap for who they are Same subject: A program that generates rich enough tests of intelligence would be as hard to build as AGI itself. — tap to centre the map on it A program that generates rich enoughtests of intelligence would be ashard to build as AGI itself. Last stated 2 years ago 6 Nov 2024 FC François Chollet — holds since 2024-11-06 — tap for who they are Same subject: Achievement is wonderful when you know why you pursue it, and a trap when you do not — tap to centre the map on it Achievement is wonderful when youknow why you pursue it, and a trapwhen you do not Last stated 11 years ago 27 May 2015 NP Natalie Portman — holds since 2015-05-27 — tap for who they are Same subject: AI is not a limited-time sale: the people furthest along are working harder than ever, and they need to stop harvesting dopamine as if it ends in two weeks. — tap to centre the map on it AI is not a limited-time sale: thepeople furthest along are workingharder than ever, and they need tostop harvesting dopamine as if itends in two weeks. Last stated 7 months ago 8 Apr 2026 DH David Heinemeier Hansson — holds since 2026-04-08 — tap for who they are Same subject: An unstarted, unresolved task is more mentally exhausting than almost any other kind of task. — tap to centre the map on it An unstarted, unresolved task ismore mentally exhausting than almostany other kind of task. Last stated 2 months ago 18 Sept 2026 GR Gretchen Rubin — holds since 2026-09-18 — tap for who they are
same subject or similar wordinga cloud: claims about one subject, named for itbar: when it was last stated, on a scale from 2015 to today — full is todaya face: someone who holds the claim — tap it for who they are

At the centre Financial incentives push public open-weight labs to make their models look better on benchmarks, even if their researchers mean to be fair. Last stated 26 Oct 2026 · 4 weeks ago Holds Nathan Lambert Read this korrent →