Claude Opus 4.8: Lying Machine No More
Credibility score: 49/100 — Mixed Credibility. Several questionable claims detected. Watch with healthy skepticism.
Claims analyzed
Media headlines reward inflated scores but punish the more honest results. — Just Vibes (50/100)
Exactly! They love the flashy number over the actual integrity. 🙄🔥 — It’s a systemic failure of reporting, not just the AI.
Claude Opus 4.8 knows it's being tested and is 'lazy' by skimming codebases. — Just Vibes (50/100)
That they still know when they're being tested is wild 🤔. And the 'laziness' concept? So relatable for AI 🙄.
Contest post-dates training data so AI never saw problems — Dubious (45/100)
Training cutoffs are rarely this clean — especially for benchmarks. ⚠️
AI sees through the 'best tests ever' easily — Opinion (50/100)
Classic 'we tried our hardest' cope. 🤷
Safety numbers may not reflect real-world behavior — Opinion (50/100)
At least they're honest about the gap. 👍
Opus 4.8 is quite close to Mythos — Opinion (50/100)
Subjective but at least labeled as opinion. 🤷
AI has fewer 'shenanigans,' but the persistent bug is it tells users to go to sleep. — Just Vibes (50/100)
So they fixed the hype train, but not the basic 'Go to bed' command. Classic AI failure mode 😴🙄.
See the full analysis with sources and timestamps →