Why Proof Convergence Matters

Why Proof Convergence Matters

🎙 Semiconductor Engineering 👥 30K 📅 April 20, 2026 ⏱ 17 min 👁 3K 📄 expert opinion 🧭 2026-08-16
Available in: English (current) Français

Keywords

formal verificationproof convergencesemiconductorAIRISC-V

Summary

In this interview, Ed Sperling, editor-in-chief of Semiconductor Engineering, discusses with Ashish Darbari, CEO of Axiomise, the critical importance of proof convergence in formal verification for semiconductor design. Darbari explains that while formal verification aims to provide deterministic yes/no answers, many properties remain inconclusive, leading to potential bugs. He highlights the increasing complexity of AI chips with hundreds of thousands of cores, which amplifies functional interactions and verification challenges. Darbari introduces Axiomise’s Core Proof technology, which uses pattern recognition and redundancy analysis to improve proof convergence and achieve sign-off quality. He emphasizes the role of AI in accelerating verification but warns about hallucination risks, advocating for formal methods to validate AI-generated results. The discussion also covers the need for industry benchmarks to standardize verification solutions and the potential convergence of formal and AI technologies in the next five years.

139 words

Critical Evaluation

Value of the Information & Strength of the Argument

The video provides valuable insights into the challenges and solutions in formal verification, particularly regarding proof convergence. Darbari’s arguments are well-structured, drawing on his extensive experience and specific examples like the hallucination incident. He effectively explains the need for deterministic answers and how Core Proof addresses this. However, the discussion remains at a high level, with limited technical detail or empirical evidence to support the claims. The argumentation is persuasive but relies heavily on anecdotal evidence and the speaker’s authority.

Scientific Rigor, Source Quality, Title Accuracy

The scientific rigor is moderate; the video is an expert opinion piece rather than a peer-reviewed study. No specific sources are cited, and the discussion lacks references to research or benchmarks. The title accurately reflects the content, focusing on proof convergence. The description mentions the impact of functional interactions and common patterns, which are discussed. However, the lack of concrete data or references reduces the overall rigor.

162 words

Title / Content Match

The title accurately reflects the content, which focuses on the importance of proof convergence in formal verification.

Quality & Reliability

7/10

The video features an expert interview with Ashish Darbari, CEO of Axiomise, providing insights into formal verification and proof convergence. The discussion is based on practical experience and industry knowledge, but lacks detailed technical depth and references to specific research or benchmarks. The claims are plausible but not independently verified.

Key Moments

Contribution & Novelties

The video offers a unique perspective on proof convergence, emphasizing the need for deterministic answers in formal verification. It introduces Core Proof technology, which uses pattern recognition and redundancy analysis to improve convergence. The discussion on AI’s role and hallucination risks provides a forward-looking view.

Pour aller plus loin :

84 words

Radar Profile

The radar profile shows balanced scores across all dimensions, with slightly lower technical depth and source rigor. This indicates a solid but not deeply technical discussion, suitable for a general engineering audience.

Reliability 7/10