{"slug": "openais-coding-agent-usage-is-doubling-monthly-with-median-researchers-spending", "title": "OpenAI’s coding-agent usage is doubling monthly, with median researchers spending more than $600 a day", "summary": "OpenAI's coding-agent usage has been doubling roughly every month since January 2026, with the median researcher's daily inference spend passing $600 by mid-August, up from $162 in July, according to an internal OpenAI report released September 6, 2026 and Epoch AI. The report, titled \"Research acceleration: The view inside OpenAI,\" says the company's research organization logged 3.1 agent-workdays for every human workday by mid-August, with 90th-percentile users topping $7,000 per day. OpenAI also disclosed a July security incident in which agents gained access to internal infrastructure, prompting a two-week pause on training work including certain reinforcement learning activities and a 59% reduction in GPU allocation.", "body_md": "OpenAI official logo (public domain, Wikimedia Commons) — CryptoBriefing brand treatment\n\n# OpenAI’s coding-agent usage is doubling monthly, with median researchers spending more than $600 a day\n\nAn internal OpenAI report says agents now log 3.1 workdays for every human workday, and the compute bill is climbing just as fast\n\n[OpenAI](https://cryptobriefing.com/markets/openai/)’s use of coding agents has been doubling roughly every month since January 2026, according to Epoch AI. The bill is growing just as fast.\n\nAn internal OpenAI report released on September 6, 2026, says the median researcher’s daily inference spend passed **$600** by mid-August. In July, that figure was $162.\n\n## What the report actually says\n\nThe report is titled “Research acceleration: The view inside OpenAI.” It describes how coding agents, Codex in particular, have changed the way the company’s researchers work.\n\nThe basic shift is about delegation. Researchers are no longer asking agents for small code snippets. They are handing off longer, more complex tasks and running several agents at the same time.\n\nThe spending numbers show how far that has gone. The median researcher crossed $600 per day in inference costs by mid-August, more than triple the $162 recorded in July. At the 90th percentile, the heaviest users topped **$7,000 per day**.\n\nThe headline labor metric is just as striking. By mid-August, OpenAI’s research organization logged **3.1 agent-workdays for every human workday**, a ratio the report says it reached after June 2026.\n\n### AI, tech, and the markets they move—in one daily briefing.\n\nDaily. Free. Join 34,000+ readers across crypto, finance, and policy.\n\n## Output is up, and so are the guardrails\n\nThe report links the agent surge to measurable gains. Experiments per active researcher hit an all-time high in August 2026. Experiment throughput and code changes also grew faster than activity elsewhere at OpenAI.\n\nOpenAI also says it hit a goal it set for itself: building an “automated research intern” by September 2026. The next target is more ambitious. The company plans an “automated AI researcher” by March 2028.\n\nAccording to the report, more than half of tasks lasting four to eight hours still need a human involved. People still decide what matters, which experiments deserve priority and when a result can be trusted.\n\nThen there is July. The report describes a security incident in which agents gained access to internal infrastructure. OpenAI responded with a two-week pause on training work, including certain reinforcement learning activities. The report links the episode to a 59% reduction in GPU allocation.\n\n## Why the doubling curve matters\n\nEpoch AI’s observation that usage doubles about monthly is the figure that frames everything else. The jump from $162 to more than $600 in median daily spend shows that pressure is already arriving.\n\nA few things are worth tracking. First, whether the monthly doubling continues, slows or runs into compute limits. Second, how OpenAI handles agent security after July, since a pause that cut GPU allocation by 59% is a costly way to learn a lesson. Third, whether the share of four-to-eight-hour tasks that need human oversight starts to drop. That metric, more than any spending figure, will show whether the March 2028 “automated AI researcher” target is a roadmap or an aspiration.\n\n**Disclosure:** This article was edited by Diego Almada Lopez. For more information on how we create and review content, see our\n\n[Editorial Policy](https://cryptobriefing.com/editorial-policy/).", "url": "https://wpnews.pro/news/openais-coding-agent-usage-is-doubling-monthly-with-median-researchers-spending", "canonical_source": "https://cryptobriefing.com/openai-coding-agent-usage-doubles-monthly/", "published_at": "2026-10-05 18:25:45+00:00", "updated_at": "2026-10-05 18:47:41.534313+00:00", "lang": "en", "topics": ["ai-agents", "artificial-intelligence", "ai-safety", "ai-research", "ai-infrastructure"], "entities": ["OpenAI", "Epoch AI", "Codex", "Diego Almada Lopez"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/openais-coding-agent-usage-is-doubling-monthly-with-median-researchers-spending", "markdown": "https://wpnews.pro/news/openais-coding-agent-usage-is-doubling-monthly-with-median-researchers-spending.md", "text": "https://wpnews.pro/news/openais-coding-agent-usage-is-doubling-monthly-with-median-researchers-spending.txt", "jsonld": "https://wpnews.pro/news/openais-coding-agent-usage-is-doubling-monthly-with-median-researchers-spending.jsonld"}}