{"schema_version":"newruntime-reader-dossiers-v1","type":"reader_dossier","id":"reader:gptbot","slug":"gptbot","name":"GPTBot","kind":"ai_training","kind_label":"AI model crawlers","operator":{"name":"OpenAI","url":"https://openai.com/"},"docs_url":"https://developers.openai.com/api/docs/bots","robots_tokens":["GPTBot"],"purpose":"OpenAI's crawler for content that may be used to train its generative AI foundation models.","feeds":["model_training"],"respects_robots":"documented_yes","robots_note":"Disallowing GPTBot signals content should not be used for training; controlled separately from OAI-SearchBot, and one crawl may serve both if both are allowed.","verification":{"method":"published_ip_ranges","url":"https://openai.com/gptbot.json"},"why_it_matters":{"text":"Training reads mean New Runtime's framing can end up baked into future OpenAI model weights — long-horizon influence, no attribution.","kind":"new_runtime_interpretation"},"confidence":"high","researched_at":"2026-10-23","sources":["https://developers.openai.com/api/docs/bots"],"icon_url":"https://newruntime.com/readers/icons/openai.png","matching":{"aliases":["gptbot"],"rule":"Reduce each agent_days label to a token: for labels starting with Mozilla/ take the product after 'compatible;', else a versioned crawler token such as 'FooBot/1.0' (a truncated 'Mozilla/5.0 (Linux; Android 7.0;) AppleWebKit/537.36 (KHTML, like Gecko) Mobile' prefix logged before 23 Oct 2026 is PetalBot; anything else is unidentified-android, unidentified-linux-chrome or unidentified-browser); otherwise take the text before the first '/', '(' or ';' and drop a trailing ' token probe'. Lowercase it. A dossier owns the tokens in its aliases."},"observed_baseline":{"as_of":"2026-10-23T08:01:01.990Z","window_first_day":"2026-07-23","window_last_day":"2026-10-23","observed_days":93,"note":"Refreshed daily in production from the live aggregates (dated by as_of). For current numbers aggregate live_sources.history agent_days by matching.aliases.","total_requests":2236,"first_seen":"2026-07-23T12:07:02.650Z","last_seen":"2026-10-22T16:25:22.427Z","days_active":43,"recency":"returning","categories":["ai_crawler"],"label_variants":1,"journeys":{"requests":10,"content_reads":0,"feeds_or_discovery_reads":2,"machine_consumption_reads":0,"unique_materials":0,"unique_canonical_nodes":0,"verification":{"claimed_unverified":{"requests":8,"method":"official_cidr"},"verified":{"requests":2,"method":"official_cidr"}}}},"identity_note":"A family name is what a client claims in its user agent. It is verified only when its request came from the operator's published IP ranges or passed forward-confirmed reverse DNS, as reported per family in /crawler-journeys.json. Anyone can claim any name.","privacy":{"scope":"family-level aggregates only","ip_addresses":"not_output","raw_user_agents":"not_output (labels are sanitized; URLs inside them are redacted)","visitor_ids":"not_output","query_strings":"not_output"},"live_sources":{"history":"https://newruntime.com/bot-traffic-history.json","latest":"https://newruntime.com/bot-traffic-latest.json","machine_consumption":"https://newruntime.com/machine-consumption.json","crawler_journeys":"https://newruntime.com/crawler-journeys.json"},"routes":{"html":"https://newruntime.com/readers/gptbot/","markdown":"https://newruntime.com/readers/gptbot.md","json":"https://newruntime.com/readers/gptbot.json","index":"https://newruntime.com/readers.json","index_html":"https://newruntime.com/readers/"}}