OpenAI's new GPT-5.4 clobbers humans on pro-level work in tests - by 83%
ID: 049329ea-7d70-5e3a-ac74-e045fe275ff1
STIX ID: report--049329ea-7d70-5e3a-ac74-e045fe275ff1
Feed Name: ZDNet Security
**Executive Summary:** OpenAI has released GPT-5.4 (the 'Thinking' model), which the company says matches or outperforms human professionals on GPTval tests 83% of the time; the article covers model availability, confusing versioning, the evaluation methodology, improvements in coding, tool use, computer vision and native computer-control capabilities, and discusses potential productivity gains and risks to jobs.
Your team is not currently subscribed to this feed. You must subscribe to it in order to see this post.