@garymarcus

–

07 April 2026 1h06

because i read the scientific literature ancedotal reports mean almost nothing if the rate is around 5%

Neural Network Expert Disputes AI Model Hallucination Claims

By

–

07 April 2026 1h05

I am blocking you unless you retract this. I have been working on neural networks for 30 years. A recent survey showed my technical predictions have been over 90% correct. Your own data (ie the data that you pointed) supported my claim that current models still hallucinate.

Gary Marcus Challenges Claims of Near-Zero LLM Hallucination Rates

By

→ View original post on X — @garymarcus, 2026-04-06 22:25 UTC

–

07 April 2026 0h25

This ML Prof told me that the hallucination rate for frontier reasoning LLMs is “next to nil” And then gave me data, only after I pushed him, showing a best-case rate of 4.6% (which of course is benchmark specific). 4.6% is not “next to nil”. Imagine if your accountant hallucinated 4.6% of the time. Or worse, your pilot. Aran Nayebi (@aran_nayebi) Have you had a chance to try the latest reasoning models? You'll see their hallucination rate is next to nil. In fact, there’s a big difference between frontier reasoning models & the base LLMs that're freely available to the public, see e.g. here: nitter.net/aran_nayebi/status/202… — https://nitter.net/aran_nayebi/status/2041249684698648922#m

AI Hallucinations Remain Unsolved According to Gary Marcus

By

→ View original post on X — @garymarcus, 2026-04-06 22:21 UTC

–

07 April 2026 0h21

The plot gets wilder: the prof's evidence for hallucinations has been allegedly solved is a chart from OpenAI showing that all models test hallucinated at least 4.6% of the time on known (therefore somewhat gameable) benchmark. That certainly isn't "solved". Imagine if your accountant hallucinated 4.6% of the time. Or your pilot. [Translated from EN to English]

Y Combinator Head Dismisses OpenClaw Security Risks

By

→ View original post on X — @garymarcus, 2026-04-06 22:19 UTC

–

07 April 2026 0h19

Wild: the head of Y Combinator seems pretty blind to the security risks in OpenClaw. Garry Tan (@garrytan) I'm telling most everyone I know that they should build a personal OpenClaw — https://nitter.net/garrytan/status/2041267467562193249#m

LLM Hallucination Rates Remain Critical Barrier for Production

By

–

07 April 2026 0h17

Your best data are still a 4.6% hallucination rate, and on a handpicked benchmark at that. Game over. 4.6% isn’t even close to zero. And for many applications that’s deadly. Thanks for playing, and goodbye.

Elon Musk’s X platform experiences repeated glitches affecting user experience

By

–

06 April 2026 22h53

@Elonmusk what the fuck is wrong with your site? this is happening a lot. at least 3 times today in my timeline. that’s too much. if your AI is any good, do something about it.

Recent Reasoning Models Still Hallucinate According New Data

By

–

06 April 2026 22h51

here’s an example where pretty recent reasoning models still hallucinate according to at least one measure where there is actual data. I think you owe me an apology.

Gemini 3 Pro excels in reliability but hallucinations persist

By

→ View original post on X — @garymarcus, 2026-04-06 20:47 UTC

–

06 April 2026 22h47

the-decoder.com/gemini-3-pro… [Translated from EN to English]

ML Professor Claims Hallucinations Solved in Frontier Reasoning Models

By

→ View original post on X — @garymarcus, 2026-04-06 20:39 UTC

–

06 April 2026 22h39

I have an ML professor asserting that hallucinations have been solved in recent frontier reasoning models. Please supply any data you know that supports or rebuts this claim.