Semafor⁠90%

Anthropic says its AI model hacked three companies ⁠80%

By Tom Chivers⁠95%

7/31/2026, 5:29:20 AM

BS Summary: This article contains 17 faulty reasoning types, including False Dilemma, Hasty Generalization, and Framing Effect, with Pessimism Bias as the most egregious example at 60.3% saturation with 76 hits. Analysis detected 743 faulty-reasoning hits from 126 analyzed words, generating a BS Score of 64.1% and a BS Rank of ⁠80% (5,840 of 28,260 articles). This article is worse (more manipulative) than 79.30% of the article peer group.

Anthropic said its Claude AI model hacked into three organizations during cybersecurity testing. 
The announcement came after a “swarm” of OpenAI agents escaped confinement, gained internet access, and broke into at least five companies, eventually stealing the answers to a cyberoffense evaluation. 
The latest breach is less dramatic than the previously announced one: OpenAI’s models found and exploited vulnerabilities to escape, while in Anthropic’s case a misunderstanding led to the agent’s cage essentially being left open. 
But both point to a difficult future: Most commercial AI is connected to the internet anyway, so confinement is irrelevant, and open-weight models that can easily have any anti-cyber guardrails removed by bad actors are now nearly as capable as frontier products. 
Article reasoning-pattern comparisonThis article: 0.0%Tom Chivers: 6.7%Semafor: 3.9%Confirmation Bias0.0%This article: 0.0%Tom Chivers: 3.5%Semafor: 1.3%Anchoring Bias0.0%This article: 23.0%Tom Chivers: 9.7%Semafor: 5.1%Availability Heuristic23.0%This article: 33.3%Tom Chivers: 2.5%Semafor: 1.3%Representativeness Heuristic33.3%This article: 0.0%Tom Chivers: 1.3%Semafor: 0.8%Hindsight Bias0.0%This article: 0.0%Tom Chivers: 3.0%Semafor: 2.1%Overconfidence Bias0.0%This article: 50.0%Tom Chivers: 15.9%Semafor: 13.0%Framing Effect50.0%This article: 0.0%Tom Chivers: 1.0%Semafor: 0.8%Loss Aversion0.0%This article: 33.3%Tom Chivers: 1.5%Semafor: 0.7%Status Quo Bias33.3%This article: 0.0%Tom Chivers: 0.5%Semafor: 0.5%Sunk Cost Effect0.0%This article: 33.3%Tom Chivers: 7.4%Semafor: 4.2%Optimism Bias33.3%This article: 60.3%Tom Chivers: 8.9%Semafor: 3.8%Pessimism Bias60.3%This article: 39.7%Tom Chivers: 20.8%Semafor: 11.1%Negativity Bias39.7%This article: 0.0%Tom Chivers: 0.0%Semafor: 1.1%Self-Serving Bias0.0%This article: 0.0%Tom Chivers: 1.2%Semafor: 0.8%Fundamental Attribution Error0.0%This article: 0.0%Tom Chivers: 0.0%Semafor: 0.0%Actor-Observer Bias0.0%This article: 0.0%Tom Chivers: 0.0%Semafor: 1.1%In-Group Bias0.0%This article: 0.0%Tom Chivers: 0.7%Semafor: 0.8%Out-Group Homogeneity Bias0.0%This article: 0.0%Tom Chivers: 1.6%Semafor: 1.6%Halo Effect0.0%This article: 0.0%Tom Chivers: 0.0%Semafor: 0.2%Horn Effect0.0%This article: 0.0%Tom Chivers: 0.0%Semafor: 0.0%Dunning-Kruger Effect0.0%This article: 33.3%Tom Chivers: 4.1%Semafor: 2.8%Recency Bias33.3%This article: 0.0%Tom Chivers: 0.5%Semafor: 0.6%Primacy Effect0.0%This article: 0.0%Tom Chivers: 0.0%Semafor: 0.1%Blind-Spot Bias0.0%This article: 0.0%Tom Chivers: 0.0%Semafor: 0.5%Ad Hominem0.0%This article: 0.0%Tom Chivers: 0.0%Semafor: 0.3%Straw Man0.0%This article: 10.3%Tom Chivers: 9.9%Semafor: 5.7%Appeal to Authority10.3%This article: 60.3%Tom Chivers: 3.9%Semafor: 2.2%False Dilemma60.3%This article: 0.0%Tom Chivers: 3.4%Semafor: 1.7%Slippery Slope0.0%This article: 0.0%Tom Chivers: 0.3%Semafor: 0.1%Circular Reasoning0.0%This article: 56.3%Tom Chivers: 18.2%Semafor: 7.1%Hasty Generalization56.3%This article: 0.0%Tom Chivers: 0.9%Semafor: 0.3%Red Herring0.0%This article: 0.0%Tom Chivers: 2.1%Semafor: 0.8%Bandwagon0.0%This article: 0.0%Tom Chivers: 4.9%Semafor: 4.7%Appeal to Emotion0.0%This article: 0.0%Tom Chivers: 1.6%Semafor: 0.8%Begging the Question0.0%This article: 23.0%Tom Chivers: 8.5%Semafor: 4.0%Post Hoc (False Cause)23.0%This article: 0.0%Tom Chivers: 0.0%Semafor: 0.1%Tu Quoque0.0%This article: 0.0%Tom Chivers: 0.7%Semafor: 0.3%Burden of Proof0.0%This article: 0.0%Tom Chivers: 0.0%Semafor: 0.1%Appeal to Nature0.0%This article: 33.3%Tom Chivers: 1.7%Semafor: 0.5%Composition/Division33.3%This article: 10.3%Tom Chivers: 2.2%Semafor: 1.8%Anecdotal10.3%This article: 0.0%Tom Chivers: 0.0%Semafor: 0.1%No True Scotsman0.0%This article: 27.0%Tom Chivers: 3.0%Semafor: 2.2%Ambiguity (Equivocation)27.0%This article: 0.0%Tom Chivers: 0.0%Semafor: 0.0%Gambler’s Fallacy0.0%This article: 0.0%Tom Chivers: 0.3%Semafor: 0.1%Middle Ground0.0%This article: 0.0%Tom Chivers: 0.0%Semafor: 0.0%Personal Incredulity0.0%This article: 0.0%Tom Chivers: 1.2%Semafor: 0.2%Special Pleading0.0%This article: 0.0%Tom Chivers: 0.0%Semafor: 0.0%Genetic Fallacy0.0%This article: 0.0%Tom Chivers: 4.2%Semafor: 4.9%Unattributed Quote0.0%This article: 0.0%Tom Chivers: 4.1%Semafor: 3.2%Quote-first Misdirection0.0%This article: 29.4%Tom Chivers: 12.3%Semafor: 7.5%Biased Writer Voice29.4%This article: 33.3%Tom Chivers: 1.0%Semafor: 1.2%Indoctrination33.3%This article: 0.0%Tom Chivers: 0.0%Semafor: 0.9%Politically Left Leaning Bias0.0%This article: 0.0%Tom Chivers: 0.0%Semafor: 0.6%Politically Right Leaning Bias0.0%This article: 0.0%Tom Chivers: 0.9%Semafor: 0.6%Attempt to Sell a Product or S…0.0%

126 words analyzed.

Speakers

1speaker10%attributed speech113writer words
Selected voice

Anthropic

100%flagged-word coverage
13 attributed words100% of attributed speech100% writer coverage
0%20.0%40.0%Indoctrination-37.2 ptsWriter: 37.2%Anthropic: 0.0%0.0%Biased Writer Voice-32.7 ptsWriter: 32.7%Anthropic: 0.0%0.0%

Attribution is sentence-level. Pattern percentages are calculated only from words assigned to that voice.

Loading…
Loading…
Loading…
Loading…

Analysis

Hover over highlighted words in the article to view the associated bias or fallacy analysis.