Open-Source Models Narrow Cybersecurity Capability Gap
The Decoder · rss · 2026-07-18
An evaluation by the UK AI Security Institute reveals that open-weight models like GLM-5.2 and DeepSeek V4-Pro now lag behind closed-source frontier models in cybersecurity capabilities by only 4 to 7 months; earlier in 2025, this gap was 6 to 10 months.
The article also notes that current safety measures for these open-source models are "largely ineffective," leaving defenders with even less preparation time. The core takeaway is that open-source models are rapidly closing the gap in cyber capabilities, but existing defenses haven't strengthened at the same pace.
More from Safety
- Ben Bajarin: Frontier AI is becoming the enterprise cyber-defense stack — BenBajarin · 2026-10-26
- DetectorScope turns black-box AI text detectors into auditable ones with sparse autoencoders — _akhaliq · 2026-10-26
- Anthropic's IPO filing devotes 94 pages to existential AI risks, from self-preservation to blackmail — Polymarket · 2026-10-26
- Study finds AI reviewing AI poorly detects reward hacking; comprehension audits proposed — ronbodkin · 2026-10-26
- A 13KB Markdown file can hijack Claude Code and Codex via your Downloads folder — pbaylies · 2026-10-26
- If an agent can rewrite itself, does your consent still mean anything? — tallmetommy · 2026-10-26