acap-foundation/acap

The ACAP project is currently a nascent proposal with no quantitative traction (0 stars, 0 forks, 0 days old). In the domain of security standards, the primary moat is not the code or the technical implementation, but industry-wide adoption and 'social gravity.' Currently, this project lacks both. It aims to fill a niche—offensive AI certification—which is distinct from the defensive focus of many existing AI safety frameworks. However, it faces intense competition from established entities like NIST (AI Risk Management Framework), MITRE (ATLAS framework), and OWASP (Top 10 for LLMs). These organizations have the institutional weight to define standards that labs and enterprises will actually follow. Furthermore, frontier labs like OpenAI (via their Preparedness Framework) and Anthropic are developing their own internal red-teaming benchmarks, which may render external 'certification' standards redundant or misaligned with the actual state-of-the-art. The project's survival depends entirely on its ability to rapidly gain contributors from the cybersecurity and AI safety communities; without that, it will likely be displaced by a standard emerging from a more established policy or security body within the next 6 months.