How to Use AI for Code Verification: Hallucinations and Formal Proof
AI assistants are changing how code gets written, but they do not make verification easier. If anything, the opposite is true. Researchers built a formally verified conference management system using AI and AutoProof for the Eiffel language. Their conclusion: AI hallucinations make correctness guarantees more important. The iterative cycle of "specify, implement, verify, fix" works even for critical systems.
AI-processed from Habr AI; edited by Hamidun News
Generative AI has helped developers: ChatGPT, Claude and other neural networks now write code from a description. But new research shows the reality: the machine handles only a small part of software engineering, while the hardest parts, namely requirements, architecture and verification, still rest with engineers.
Can a neural network fully replace a programmer?
No. Developers spend 20 to 30% of their working time on actual coding. The rest goes to gathering requirements, design and verification, which call for human experience.
- Nov 1, 2026Local models lose to Opus and Codex: 0.21 on a MacBook versus 1.00 for Codex
- Nov 1, 2026Kimi scores 96.2% on a 1C database analytics test and emerges as a fallback if Claude access is blocked
- Nov 1, 2026OpenAI Rewrites GPT Behavior Rules: Five Values and a Chain of Command
- Nov 1, 2026Google admits AI Mode cuts search revenue, shares fall 15%
Test Frontier LLMs & Autonomous Agents in One Workspace
Run Claude Opus 5.5, GPT-6, Gemini 3.1, and DeepSeek side-by-side in Nomad AI (nomadnet.ai) — unified model routing, deep research, and multimodal workflows under one subscription.
The AI world, distilled — once a week
Seven stories that actually mattered, hand-picked. No noise, no reposts, no press releases.
Done! Check your inbox for a confirmation.