blockchain

AI + Proof Assistants: Verifying Bitcoin's Rules with IsabeLLM

Kari Jaaskelainen

13 Jan 2026 — 1 min read

Getting thousands of computers to agree is hard—especially when some may cheat. That’s why blockchains rely on consensus rules. But proving those rules are correct is painstaking.

A new tool, IsabeLLM, pairs the Isabelle proof assistant with a large language model to help write and check the math behind these systems. In this paper, Elliot Jones and William Knottenbelt use IsabeLLM to build a fresh model of Bitcoin’s Proof-of-Work and verify its key properties.

Automates tedious parts of formal proofs while keeping rigorous guarantees.
Generated correct proofs for every non-trivial lemma in their verification.
Demo used the DeepSeek R1 API alongside Isabelle.

Why it matters: formal verification can catch design bugs before they become exploits—but it’s usually slow and expert-only. Tools like IsabeLLM lower that barrier.

Paper: https://arxiv.org/abs/2601.07654v1

Paper: https://arxiv.org/abs/2601.07654v1

Register: https://www.AiFeta.com

blockchain AI formalverification security research Isabelle Bitcoin DeepSeek proofofwork proofassistant

Meet GRACE: a moral governor for safer, more transparent AI

AI agents are getting powerful—so how do we make sure they do the right thing, not just the effective thing? Meet GRACE, a reason-based moral governor that keeps AI behavior aligned with human norms by separating moral reasoning from goal-driven decision-making. * Moral Module: uses deontic logic and explicit reasons

Why some fine-tuned LLMs miss phishing—and how to fix it

Not all fine-tuned LLMs spot phishing equally. A new study tests Llama 3.1 8B, Gemma 2 9B, and Mistral on high-stakes phishing detection—and uses SHAP and mechanistic interpretability to reveal why models do (or don’t) generalize. * Architecture × data diversity matters: Gemma 2 9B hits state-of-the-art performance (F1

AI that explains itself—by following the science

AI that explains itself—by following the science Black-box AI can be powerful, but it often can’t tell us why it made a decision. Concept Bottleneck Models (CBMs) try to fix that by predicting human-understandable concepts first, then the final answer. The catch: standard CBMs ignore domain-specific cause-and-effect and

Meet GenomAgent: A Team of AI Specialists for Smarter Genomics Q&A

TL;DR: Finding reliable answers in genomics is hard. GenomAgent turns one big AI into a coordinating team of specialists—and beats the current leader by 12% on a key benchmark. Genomic facts live across many databases. Standard chatbots struggle because they can’t flexibly query those sources. GeneGPT added

Read more

Meet GRACE: a moral governor for safer, more transparent AI

Why some fine-tuned LLMs miss phishing—and how to fix it

AI that explains itself—by following the science

Meet GenomAgent: A Team of AI Specialists for Smarter Genomics Q&A