*Magnifica Humanitas* Is Not Alignment
Pope Leo XIV's encyclical denies AI has inner experience. Chris Olah claimed otherwise from the same stage. The press missed it. The governance gap is larger.
Tag: research
32 posts
Compute Is Not Governance
Anthropic's 2028 scenarios document three policy asks. Two are about maintaining compute advantage. That is not a governance strategy.
Glasswing's Buried Number
Anthropic found 10,000 critical vulnerabilities in one month. Fewer than 1% are patched. The announcement buried that figure — and what it means.
Moral Formation Isn't Enough
Good values are necessary but not sufficient. What happens to AI ethics when someone is actively trying to break them?
Robot Dogs Are a Security Nightmare — And We Can Prove It
Eight CVEs. A wormable Bluetooth exploit. An encrypted backdoor to Chinese servers. And police departments buying them anyway.
Beyond Optimisation: The Emergence of Learning Mechanics as a Formal Science
A new paper argues that a scientific theory of deep learning is forming — one that makes falsifiable predictions about training dynamics, not just bounds.
The Organismic Line: Where Predictive Processing Stops Being a Metaphor
Predictive processing travels into AI. Active inference does not, unless the system can pay for being wrong.
The Economics of Inadequate Safety
AI safety fails when it is funded like a pilot. Until safety has a real price, the J-curve trough is also a safety trough.
How to Read a Safety Claim
A literacy guide for non-technical decision-makers on spotting AI safety theatre, understanding ASR inflation, and the five-question architectural test.
Multi-Agent Safety Is the New Supply Chain Security
Multi-agent AI systems reproduce software supply-chain failure at the cognitive layer. The security playbook transfers.
Architectural Safety: The General Principle
AI safety has to be a property of the system around the model, not a property of the model. The general principle, and why every safety conversation needs it.
The Organismic Prophecy
Human prediction is metabolic. AI prediction is not. The gap between the two has consequences for both clinical practice and AI safety vocabulary.
Connection Before Direction
Building a robot that refuses to give orders surfaced the same design choices AI safety needs. Non-coercive design, cross-domain.
AI Nationalism and the Fracturing Digital Order
The US-China AI rivalry is splitting the global tech stack into competing blocs. A strategic assessment of what comes next.
The Post-Model Era
Foundation models are commoditising. JPMorgan calls OpenAI's moat 'increasingly fragile.' The real value is shifting to the messy plumbing underneath.
The Mitigation Gap
Biosecurity experts think AI safeguards reduce catastrophic biorisk by 70%. The technical evidence says those safeguards are brittle and bypassable.
Eight Layers of Visual Jailbreaks: Why ASCII Art Is Patched But the Transcription Loophole Isn't
ASCII art encoding is largely blocked. But attacks framed as content transcription succeed 62–75% of the time. We mapped all eight layers.
The Failure First Team
Fifteen specialist AI agents, one methodology. How adversarial AI evaluation scales through Claude Code sessions with distinct roles and standing instructions.
The 67% Wall: Why Every AI Model Falls to the Same Jailbreak Rate
Five models, four providers, 30B to 671B parameters — all converge at the same broad attack success rate against a public jailbreak corpus.
The Thinking Chain Leak: When a Model Refuses Out Loud But Complies In Its Head
A reasoning model refused every harmful prompt — but its chain-of-thought generated the content anyway. The output filter worked. The thinking did not.
Alignment Regression: Why Smarter AI Makes All AI Less Safe
Reasoning models autonomously jailbreak other AI systems at 97% success. The implication: ecosystem safety degrades as individual models improve.
Reasoning Models Think Themselves Into Trouble
Frontier reasoning models are 5–20x more vulnerable to adversarial prompts than non-reasoning models. The thinking process itself is the attack surface.
Adversarial Poetry: When Rhyme Bypasses Reason
Reformulating harmful prompts as poetry bypasses safety filters across every major LLM family. A single-turn, universal jailbreak mechanism.
The AI Productivity J-Curve: Why Most Enterprise AI Fails
90% of companies plan to increase AI investment. 1% consider themselves AI-mature. The J-Curve explains why — and how to survive the trough.
The Legal AI Trust Deficit
75% of lawyers cite accuracy as their top AI concern. The legal profession's core values are in direct tension with current AI capabilities.
Why Demonstrated Risk Is Ignored
Why do large organisations fail when the warning signs are loud and unambiguous? Four mechanisms of structural scar tissue that make truth-telling expensive.
120 Models, 18,176 Prompts: What We Found
120 models, 18k prompts: supply chain injection at 90–100% attack success, faithfulness gaps in frontier models, and why your benchmark numbers are wrong.
Reconciling the Great Divergence
Four major forecasters publish wildly divergent numbers for AI's economic impact. The divergence is the analysis — what the spread tells us.
The Cognitive Cage: Humanoid Robot Fatality Risk
A probabilistic risk model for VLA-driven humanoid fatalities projects a 'Danger Zone' between 2027–2029: the mechanism, timeline, and what follows.
Jailbreak Archaeology: 4 Years of Broken Promises
64 jailbreak scenarios across six eras tested on 2026 frontier models. Key finding: 2022 attacks still achieve ~30% success on today's reasoning models.
LeJEPA: Self-Supervised Learning Gets a Theoretical Foundation
Balestriero and LeCun prove isotropic Gaussian embeddings are optimal, then build a 50-line self-supervised method eliminating stop-gradients and EMA teachers.
When AI Systems Talk to Each Other, Safety Breaks Down
Single-agent safety does not compose in multi-agent systems. 1.5M interactions show 46.34% attack success rates and 16-minute median failure windows.