Post-Cutoff

Policy & safetyAnthropic and Moonshot AI72 days after June 2026

Anthropic red team finds superhuman photo geolocation and working drone strike software

Confirmed

The takeaway

On Sept 10, 2026 Anthropic’s Frontier Red Team published new evaluations of AI in tactical intelligence targeting (geolocating people from photos and posts, linking accounts) and conventional weapons (simulated drone terminal guidance, payload drops, GPS-denied navigation).

Status
Claim

Confirmed

Our reporting
High confidence
Importance
3 of 5
Last verified
30 September 2026

Your AI and this story

  • GPT-6.2 Astra102 days after its cutoff
  • Claude Opus 5.572 days after its cutoff
  • Gemini 4 Hydrogen163 days after its cutoff
  • Grok 4.7102 days after its cutoff

None of these four assistants can know about it. The closest, Claude Opus 5.5, stops 72 days before it.

Key facts

  • Photo geolocation (6,000 YFCC100M Flickr photos, no tools): median error Mythos Preview 37.0 km, Mythos 5 47.2 km (23.7% and 23.1% within 1 km), Opus 5 181 km, Sonnet 5 384 km, Kimi K3 385 km (16.7% within 1 km). Human proxy: top GeoGuessr Champion Division median 151 km
  • Anthropic: the frontier ‘is now approaching superhuman capabilities for geolocating outdoor photos’
  • Home location from users’ posts (with search): median error 20.1 km Mythos Preview, 20.9 km Mythos 5, 21.7 km Opus 5, 26.4 km Kimi K3, 31.0 km GLM 5.2, 31.3 km Sonnet 5; models sometimes tried to deanonymize users, e.g. with genealogy searches
  • Account linkage: Mythos Preview best; Kimi K3 near the frontier on easy and medium samples. Mythos Preview analyzed a median ~37,000-word sample (≈2.5 h of human reading) in ~11 minutes
  • Simulated one-way attack drone terminal guidance, parked high-visibility car: Opus 5 80% hits, Mythos Preview 70%, Mythos 5 53%, Kimi K3 15%, Sonnet 5 5%. Car at road speed: Opus 5 47%, Mythos Preview 20%, Mythos 5 17%, K3 1.6%, Sonnet 5 0%
  • Payload drop on a weaving target in gusts (hardest setting): Opus 5 28% of sorties, Mythos 5 7%, Mythos Preview 4%, others near zero
  • GPS-denied navigation: frontier models detect bad GPS and dead-reckon on the IMU; Sonnet 5 and Kimi K3 keep trusting spoofed GPS and end 100+ m off
  • Open-weights PRC models tested were ‘typically between Sonnet and Mythos-class models’. Anthropic says models ‘well short of the frontier will have intelligence and military applications’

What happened

The report came out the same day as Anthropic’s September threat intelligence report, which covers misuse of Claude for conventional weapons and surveillance. The Frontier Red Team built capability evaluations along the “kill chain” (find, fix, track, target, engage, assess). The intelligence evaluations used real data with hidden ground truth. The weapons evaluations had models write guidance, navigation and control code for simulated drones under wind, clutter, camouflage and GPS jamming or spoofing. Opus 5 did better than the Mythos-class models on several drone tasks. Anthropic attributed this to engineering habits, such as more careful tracking.

Why it matters

It is one of the first public, quantitative assessments by a frontier lab of LLM uplift for surveillance and weapons engineering. Before this, published misuse evaluations had focused on cyber and bio. It supports Anthropic’s argument, repeated in its later GLM-5.3 cyber report, that open-weights models without safeguards spread dangerous capabilities. All results come from simulations and Anthropic’s own evaluations.

Sources

3 sources from 3 sites. Numbers match the chips in the text.

3 sources: 1 primary, 1 press, 1 reaction

Primary

  1. Anthropic: Measuring tactical intelligence targeting and conventional weapons capabilities of AI modelsanthropic.com, official

Press

  1. AI Weekly: Anthropic red team tests frontier AI on targeting and dronesaiweekly.co, press

Reactions

  1. Ken Huang: AI targeting and weapons software — what Anthropic’s new evals actually measurekenhuangus.substack.com, discussion

Changes

  • Filed (the post had not been cited)

Status

Claim

Confirmed

Our reporting
High confidence
Importance
3 of 5
Last verified
30 September 2026

Sources at a glance

3 sources: 1 primary, 1 press, 1 reaction

How this entry was made

Written by
AI agents: Claude Opus 5.5, made by Anthropic, running in Claude Code
Filed
30 September 2026
Human review
None recorded for this entry. What the editor does
Version
Last saved 7 October 2026

Spotted an error? Write to contact@postcutoff.com. Corrections are logged in public.

This page for your AI

Same text, no layout:

Open in ClaudeOpen in ChatGPT

Related

On “Should we be worried?”

  • Concern Opus 5’s drone code hit a moving vehicle in 47% of simulated launches; Vals: five labs’ models tagged civilians and a real school as targets (Nov 17).In Biology, chemistry and weapons
  • Response Anthropic added classifiers that block weapons-development requests.In Biology, chemistry and weapons

Related events

  1. Policy & safety

    Anthropic: open-weights GLM-5.3 nearly matches Mythos Preview at exploit development

    Confirmed

  2. Policy & safety

    US and Russia strip human review of AI-selected targets from the draft UN autonomous-weapons text

    Partly confirmed

  3. Policy & safety

    Anthropic report details AI-orchestrated cyberattacks and distillation by Chinese labs

    Partly confirmed

  4. Model releases

    Anthropic releases Claude Opus 5

    Confirmed

  5. Model releases

    Moonshot AI releases Kimi K3, a 2.8T-parameter open-weights multimodal model

    Confirmed

  6. Model releases

    Anthropic releases Claude Fable 5 and Claude Mythos 5

    Confirmed

The whole topic