corpus.blog
43,567 blogs · 5,563,491 posts Claim your blog
Record · corpus.blog/posts/01a0f789-b41c-70ba-87b4-73d611edf691

Why AI benchmarks are broken

bdtechtalks.com · published 15 December 2025

A record is what was published and who pointed at it. The text of the post is not held here: read it at the source, or at its Wayback capture.

Read the post at bdtechtalks.com Wayback capture
0blogs citing
First cited
—
Most recent
—
Rank this month
not ranked
External links
18
Words captured
1,011

Cited by

No blog we hold has cited this post yet.

Links from this post

Link
Host
https://i0.wp.com/bdtechtalks.com/wp-content/uploads/2025/12/LLM-benchmark-race.jpg?ssl=1
i0.wp.com
https://x.com/arcprize/status/1999182737107042736
x.com
https://artificialanalysis.ai/
artificialanalysis.ai
https://x.com/EpochAIResearch/status/1999548496198926728
x.com
https://simple-bench.com/
simple-bench.com
https://t.co/X6JnJuDuRz
t.co
https://twitter.com/kyleichan/status/1999292461450166350?ref_src=twsrc%5Etfw
twitter.com
https://www.theregister.com/2025/04/08/meta_llama4_cheating/
theregister.com
https://twitter.com/bendee983/status/1999149000193196403?ref_src=twsrc%5Etfw
twitter.com
https://en.wikipedia.org/wiki/Goodhart%27s_law
en.wikipedia.org
OpenAI GDPval Read at the source ↗
openai.com
FACTS Benchmark Suite: Systematically evaluating the factuality of large language models Read at the source ↗
deepmind.google
https://twitter.com/bendee983/status/1999147703754731951?ref_src=twsrc%5Etfw
twitter.com
https://en.wikipedia.org/wiki/Catastrophic_interference
en.wikipedia.org
https://x.com/karpathy/status/1816531576228053133?lang=en
x.com
https://t.co/3C7pCdBShQ
t.co
https://twitter.com/karpathy/status/1816531576228053133?ref_src=twsrc%5Etfw
twitter.com
https://t.co/Vtd848HNER
t.co
Elsewhere on bdtechtalks.com (4)