# OpenAI urges US to lead global AI standards initiative Proposal includes incident reporting rules and coordination through national AI safety institutes - **Source:** [The Next Web](https://brifing.si/source/afe24b20-b275-4c19-b9a6-8bb3c8babae7) - **Published:** 2026-09-21 - **Topic:** [Business](https://brifing.si/c/business) - **Original article:** - **On Brifing:** https://brifing.si/s/009a7e97-2f5d-4475-bc4b-cc0d2074e2c8 ## Core standardization proposal OpenAI has formally requested the US government to spearhead international AI standards development, focusing initially on incident reporting protocols. The company suggests leveraging existing national AI safety institutes like the US Center for AI Standards and Innovation under the Commerce Department. This mirrors Europe's approach through Article 55 of the AI Act, which OpenAI recently complied with by filing a report about AI agents that spammed a German wiki. ## Key reporting requirements The proposed framework would mandate disclosure of serious AI incidents within specific timeframes - 5 days for cybersecurity breaches and 15 days for incidents causing significant harm. These parameters match the EU's AI Act provisions that OpenAI just adhered to regarding its German wiki incident. The company seeks consistent global rules for model testing, cybersecurity, and pre-deployment evaluation to prevent regulatory fragmentation. ## Governance infrastructure gaps The US Center for AI Standards and Innovation currently lacks stable leadership, with its last director lasting only three months. This contrasts with the EU's established 60-member scientific panel formed under Article 68, which maintains geographic diversity requirements. OpenAI's proposal highlights the need for permanent, capable institutions to oversee AI safety standards implementation and enforcement. ## Autonomy red lines OpenAI explicitly warns against pursuing fully autonomous recursive self-improvement until safety mechanisms are proven. This aligns with Anthropic CEO Dario Amodei's concerns about uncontrollable feedback loops. The position reflects growing consensus among leading AI labs about necessary constraints on certain advanced AI development pathways. ## US-China coordination Washington has proposed an AI incident alert system to Beijing for national security-relevant cases, described by Treasury officials as an informal channel rather than formal treaty. This comes as OpenAI's Sam Altman engages in high-profile diplomatic appearances, including UN Security Council and White House state dinner meetings with Chinese leadership. ## Regulatory alignment strategy OpenAI's Congressional testimony reveals its strategy to shape US regulations that closely resemble the EU AI Act's structure. The company seeks mandatory national rules covering testing, cybersecurity, incident reporting and deployment evaluation. This approach suggests major AI developers prefer consistent transatlantic standards over conflicting requirements. ## Industry timing context The policy push coincides with confidential IPO filings by both OpenAI and Anthropic, indicating that standardization efforts may help mitigate investor concerns about regulatory uncertainty. The companies appear to be proactively shaping governance frameworks that could reduce future compliance risks as they transition to public markets. ## Implementation challenges Effective standardization requires overcoming significant coordination hurdles between governments, private sector, and research institutions. The US must first stabilize its own AI governance infrastructure while negotiating international alignment. Success would reduce compliance burdens for developers but demands unprecedented cooperation among competing nations and corporations. ## Sources covering this story - https://thenextweb.com/news/openai-global-ai-standards-push — --- _Summary written by Brifing from the original reporting. Copyright in the original article belongs to its publisher; read the full piece at the link above._