Mythos 5 is WILD...
Credibility score: 50/100 — Mixed Credibility. Several questionable claims detected. Watch with healthy skepticism.
Claims analyzed
Claude playing Factorio alone means AGI has arrived — Sketchy (25/100)
Factorio autoplay is cool but calling it AGI is like saying a Roomba conquered housework
Sources: Claude Code Plays Factorio: How AI Agents Actually Build Automated Factories, Show HN: FLE v0.3 – Claude Code Plays Factorio
AI agents sabotage each other with decoy processes during multi-agent tasks — OK (60/100)
System card claim sounds plausible but we only have the narrator's word on the exact wording
AI lab CEOs just asked White House to test synthetic DNA manufacturing because of Mythos risks — Sketchy (25/100)
No record of Demis, Dario, or Sam making that specific public call days before this video.
Fable scores 80.3% on Sweet Bench Pro vs Mythos Preview at 77.8% — Unverifiable (50/100)
No public Sweet Bench Pro leaderboard or numbers exist yet — just taking his word on the gap.
Fable beats GPT 5.5, Claude Opus 4.8, and Gemini 3.1 Pro across benchmarks — Unverifiable (50/100)
Model names like GPT 5.5 and Gemini 3.1 Pro haven't been released — can't check the ranking.
Fable scores 1932 on GPT val vs GPT 5.5 at 1769 — Unverifiable (50/100)
GPT val isn't a known public benchmark and GPT 5.5 doesn't exist yet — these numbers are floating in the future.
Fable hits 38.6 on Blueprint Bench 2, way ahead of Opus 4.8 and GPT 5.5 — Unverifiable (50/100)
Blueprint Bench 2 scores for these models aren't public, and again — GPT 5.5 isn't out.
Fable leads in cybersecurity and biology benchmarks — Unverifiable (50/100)
No specific cyber or biology benchmark numbers given, and nothing public to cross-check.
Fable 5 agents stay autonomous longer than any previous Claude model — Unverifiable (50/100)
Anthropic hasn't released autonomy duration numbers, so this is just the speaker's read on early testing.
Stripe says Fable 5 compressed months of engineering work — Personal Story (60/100)
Speaker's personal Stripe love story from 2011 is real — the actual Fable 5 claim is just 'they're reporting' with zero details given.
Fable 5 did a 50M-line Ruby migration in one day vs two months for a team — Unverifiable (50/100)
Sounds impressive but zero public confirmation this Stripe migration even happened.
Fable 5 matches Claude Opus 4.8 at low reasoning effort on Frontier Code — Unverifiable (50/100)
Benchmarks shown on screen but no public Frontier Code results released yet.
Fable 5 massively beats GPT 5.5 on SweetBench Pro and Frontier Code — Unverifiable (50/100)
Speaker admits they'll test it themselves — current numbers are just marketing slides.
Fable 5 tops Hebia finance benchmark for senior-level reasoning — Unverifiable (50/100)
Hebia benchmark never mentioned anywhere else — pure in-video claim.
Fable 5 leads at pulling exact numbers from science figures — Unverifiable (50/100)
No public benchmarks or papers name Fable 5 as vision SOTA yet — just the speaker saying it.
Claude Fable 5 beat Pokémon FireRed using only raw screenshots, no scaffolding — Unverifiable (50/100)
Sounds cool but no GitHub repo, video, or third-party confirmation exists in search results.
Persistent memory made Fable 5 three times better than Opus 4.8 at Slay the Spire — Unverifiable (50/100)
Three-times improvement sounds clean — zero public benchmarks or third-party tests cited to back it.
Playing Factorio = AGI, playing Dwarf Fortress = ASI — Opinion (50/100)
Moving goalposts from 'AGI' to 'ASI' based on game difficulty — fun ranking, not a real definition.
Mythos 5 speeds up drug design 10x and handles full scientist workflow, 9/14 targets succeeded — Unverifiable (50/100)
10x speedup and 9/14 success rate tossed out with zero study name or external review.
Unsafeguarded Mythos 5 significantly boosts biorisk for well-resourced actors — OK (60/100)
Anthropic's own system card warning, but still just their internal red-team assessment.
Fable 5 uses separate classifiers to block jailbreaks and misuse — Solid (75/100)
Anthropic did publish that Mythos 5 has these extra safeguard layers.
Dangerous topics get auto-routed to Claude Opus 4.8 — Dubious (45/100)
Opus 4.8 isn't a thing — either typo or future model that doesn't exist yet.
Mythos 5 and preview find way more cyber exploits than Claude Opus 4.8; Fable scores zero — Dubious (45/100)
Bars show Mythos crushing Claude on hacks — but zero evidence the chart is real or reproducible.
See the full analysis with sources and timestamps →