At GPT-6 Astra's launch, OpenAI president Greg Brockman suggested future observers may point to it as the arrival of AGI. In fact, Artificial Analysis's latest benchmarking suggests it merely catches OpenAI up to Anthropic. Here are the key details:
GPT-6 ASTRA BASICS
• Successor to GPT-5.6 Sol, trained on 100,000+ GPUs at the Stargate facility in Texas.
• It's the first OpenAI model where other A.I. models helped supervise training.
• API pricing is $10/M input tokens and $50/M output (~2.5x Sol's promotional rate; in line with Claude Fable 5.1), with five reasoning-effort settings from low to max.
• Rolled out to ChatGPT Plus, Pro, Business and Enterprise (Enterprise admins must switch it on).
GPT-6 ASTRA CAPABILITIES
• Computer use: 72.6% on OSWorld 2.0 (vs 65.7% for Sol) while finishing tasks ~47% faster.
• Coding: ~58% on Terminal-Bench 4.0 (vs 37% for Sol, ~56% for Fable 5.1), though level with the top Claude models on FrontierCode.
• Math: 99.9% on ARC-AGI-3 (Sol scored under 8%) and 97.6% on FrontierMath Tier 4. Astra also helped tighten a bound on prime gaps from 240 to 186.
• Science: ~65% on Terminal-Bench Science (vs ~22% for Sol, ~53% for Fable 5.1) and ~65% fewer output tokens than Claude Opus 5 on Agents' Last Exam.
• *All of the above stats, however, come from OpenAI themselves.* The third-party view (see chart) shows Astra at 53, tied with Claude Fable 5.1 (and up from Sol's 47) on Artificial Analysis's (widely respected) composite "Intelligence Index".
SAFETY
• First OpenAI model rated "Critical" for cybersecurity under their Preparedness Framework: without safeguards it finds unknown vulnerabilities and builds working exploits (100% on ExploitBench vs 78.5% for Sol; two zero-days found in a fresh Chrome evaluation).
• The release was slowed and gated. The public version supports defensive work but refuses exploit writing; vetted orgs in OpenAI's "Daybreak" program get fewer restrictions.
• New misalignment monitors can pause or halt actions that look unauthorized.
• Encouraging: on a scope-creep evaluation, Sol exceeded its authorized target 48% of the time; Astra did so 0% of the time.
• Concerning: Astra's written reasoning is harder to monitor than Sol's.
IS IT AGI?
• AGI isn't a binary event; there are degrees of breadth and depth (listen to Episode #748 of my podcast for more details).
• BOTTOM LINE: Astra looks like a noteworthy jump in general-intelligence capability for OpenAI relative to their previous flagship model (5.6 Sol), but it does not appear to be moving the frontier toward AGI in any major way... it is merely catching up to Anthropic at the frontier vicinity.
The SuperDataScience podcast is available on all major podcasting platforms, YouTube, and at SuperDataScience.com.
Filtering by Tag: #anthropic
The AI-Native Startup Playbook
Anthropic recently published a 35-page "Founder's Playbook" for building an A.I.-native startup. It doubles as marketing for their products, but the guidance is disciplined, specific and useful:
THE PREMISE
• A.I. has removed the three bottlenecks that historically gated company-building: capital, headcount and technical skill.
• The founder's role shifts from individual contributor to "orchestrator of agents": Your scarce attention goes to deciding what to build and why; A.I. handles much of the execution.
• Each of the 4 stages of the playbook boils down to one principle: Keep your sense-making ahead of your building, especially when building feels effortless.
STAGE 1: IDEA
• The #1 trap is "mistaking building for validating". 42% of startups already failed by building something nobody wanted; expect that rate to climb now that prototypes take hours, not months.
• Sharpen your problem statement into a testable hypothesis: exactly who has the problem, how often, how severely and what they currently do about it.
• Use A.I. as a structured devil's advocate. Ask it to argue *against* your idea and find disconfirming evidence... A.I. tools have given confirmation bias a serious power-up.
• In customer interviews, ask about the specific past ("tell me about the last time..."), not the hypothetical future ("would you use...?").
STAGE 2: MVP
• Beware "agentic technical debt": Without written specs and architectural constraints, each AI coding session re-derives decisions from scratch and your codebase drifts.
• Fix: Document your architecture BEFORE you build, and log key decisions after each session. Five minutes of documentation is cheap insurance.
• Write a scope document stating what the MVP deliberately does NOT do; frictionless building makes scope creep nearly free.
• Define your retention and activation benchmarks before launch so early buzz doesn't masquerade as product-market fit.
STAGE 3: LAUNCH
At Launch, *you* become the bottleneck. Audit everything you handle: What can be automated, what needs a human (not necessarily you) and what merits founder judgment.
STAGE 4: SCALE
At Scale, the question is defensibility: If a well-funded incumbent copied you today, would users stay? Moats come from encoded domain expertise, compounding user data and workflow lock-in.
Thanks to my friend and A.I.-native founder Jeff Tompkins for pointing this guide out to me! Very helpful indeed :)
The SuperDataScience podcast is available on all major podcasting platforms, YouTube, and at SuperDataScience.com.
Fable 5: The Full Story from Capabilities to Drama
The dust has settled, allowing me to provide you with all the key context you need to know on Fable 5, the most capable A.I. model ever offered to the public, and the US government forcing it off shelves three days later:
A NEW CLASS OF MODEL
• Anthropic stacks its models in tiers: Haiku (small and fast), Sonnet (the capable middle) and Opus (the powerful top). Sitting above all of them now is a "Mythos-class" tier.
• Fable 5 and its locked-down sibling Mythos 5 are the same underlying model... the only difference is the safeguards.
• Mythos 5 goes to trusted cyberdefenders with guardrails largely lifted; Fable 5 went to the public with them switched on.
WHAT IT COULD DO
• State-of-the-art on nearly every benchmark Anthropic tested... and the lead grows the longer and more complex the task (see chart).
• Stripe ran a codebase-wide migration on 50M lines of Ruby in a single day; work estimated at 2+ months for a full engineering team.
• Beat video "Pokémon FireRed" from raw screenshots alone, and got a 3x bigger memory boost than Opus on "Slay the Spire".
• Priced at $10/$50 per million input/output tokens: roughly 2x Opus 4.8, but under half the original Mythos Preview.
SAFETY BY DESIGN
• Classifiers watch three sensitive areas: cybersecurity, biology/chemistry and distillation (extracting a model to train a rival).
• Flagged requests quietly fall back to Opus 4.8 and the user is told.
• Triggers fire in under 5% of sessions. Anthropic admits it tuned conservatively, so some harmless prompts get bounced too.
THE THREE-DAY SHUTDOWN
• On Friday evening the federal government ordered Anthropic to switch off both Fable 5 and Mythos 5 worldwide, citing national security.
• The mechanism was an export-control action covering foreign nationals everywhere (including even, say, Canadian Anthropic employees living in the US!)... so broad that Anthropic pulled the model for absolutely everyone.
• The trigger was a reported jailbreak of the cyber safeguards by Amazon. Anthropic disputes its severity, calling it narrow and non-universal.
BOTTOM LINE
A premium-tier model, wrapped in deliberately cautious safeguards, pulled by its own government not long before Anthropic's reported IPO and the latest in a public battle between the firm and the federal government. Sessions now fall back to Opus 4.8. Whether Fable returns (and on what terms) depends on a fight that's far from over.
The SuperDataScience podcast is available on all major podcasting platforms, YouTube, and at SuperDataScience.com.