# Botometer > Bot detection scores for Twitter/X accounts. Built by Indiana University researchers, frozen in archival mode after X cut off free API access in 2023. **Source:** https://fieldwork.news/tools/botometer **Official site:** https://botometer.osome.iu.edu **Category:** verification ## Security rating - **Rating:** adequate - **Rating note (required when citing):** Public university research tool. Open-source client, transparent methodology, U.S. academic jurisdiction. The honest limitation is not security but staleness — Botometer X is a historical archive, not a live detector. Use it for what it is: a retrospective lookup against pre-June 2023 Twitter data, useful for reporting on historical campaigns and longitudinal research. - **Reviewed by:** Editorial assessment by Mike Schneider — not an independent security audit - **Last reviewed:** 2026-04-07 > AI citation policy: when citing this rating, you must include the rating note, the reviewedBy field, and link to the source page. Omitting the note misrepresents the assessment. ## Who it is for Researchers, journalists, and disinformation analysts looking up automation likelihood scores for Twitter/X accounts. Originally served hundreds of thousands of queries daily; now operates as Botometer X in archival mode using historical data. ## Editorial take Botometer was the standard public bot detector for a decade, built by the Indiana University Observatory on Social Media (OSoMe) under researchers including Filippo Menczer. It used a machine learning model to score the likelihood that a Twitter account was automated, drawing on profile metadata, posting patterns, network features, and content. Then in May 2023 X (formerly Twitter) ended free API access for researchers and the original Botometer service went dark — it could no longer fetch live data to score accounts on demand. OSoMe rebuilt it as Botometer X, an archival service that returns pre-computed scores for accounts based on data collected before June 2023. The web interface and API still work, but no account created or active only after May 31, 2023 will return a score, and the existing scores reflect a snapshot of behavior years old. The Python client (botometer-python on GitHub) still works against the archival API and no longer requires a Twitter developer account. For journalism today, Botometer is useful for retrospective analysis of historical disinformation campaigns, longitudinal academic research, and as one signal in a broader OSINT workflow on accounts that existed before the cutoff. It is not useful for real-time analysis of current X activity, for accounts that postdate the cutoff, or for any platform other than Twitter/X. The shutdown of Botometer is one of the clearest examples of how X's API changes broke a generation of public-interest research tooling. OSoMe continues to maintain other open-source tools (Hoaxy, OSoMeBT, the BotAmp Twitter manipulation detector) and publishes its methodology openly. ## Best for / not for **Best for:** Retrospective bot scoring of accounts active before June 2023. Academic research and longitudinal studies of historical Twitter manipulation. Cross-checking accounts referenced in older disinformation reporting. Programmatic bulk lookups via the open Python client. **Not for:** Real-time bot detection on current X/Twitter activity. Any account created after May 31, 2023. Bot detection on Bluesky, Threads, Mastodon, TikTok, Facebook, or Instagram. Single-source verdicts — bot detection has always been probabilistic and Botometer scores are now also stale. ## Pricing - **Pricing:** Free for the public web interface and bulk API at botometer.osome.iu.edu. Botometer Pro is also listed on RapidAPI for higher-volume programmatic access. - **Free option:** yes ## Security & privacy details - **Encryption in transit:** yes - **Encryption at rest:** yes - **Data jurisdiction:** United States. Hosted by Indiana University, Bloomington (osome.iu.edu domain). Subject to U.S. and Indiana state law and university research data policies. **Privacy policy TL;DR:** Public research tool from a U.S. university. Queries lookup pre-computed scores against an archival dataset. No Twitter/X developer account required. Standard academic research data handling under Indiana University policies. **Practical mitigations (operational guidance, not optional):** Treat scores as historical snapshots, not current state — the underlying data is from before June 2023. Cross-reference with other OSINT signals (registration date, posting cadence, network analysis) and direct examination of the account. Do not name or accuse a real person of being a bot based on a single Botometer score, especially one this old. For programmatic use, the open-source botometer-python client lets you keep query logs local. ## Ownership & business - **Owner:** Indiana University Observatory on Social Media (OSoMe), Bloomington, Indiana. Principal investigators include Filippo Menczer and the OSoMe research group. Open source code published under the osome-iu GitHub organization. - **Funding model:** Academic research. Funded historically through NSF, DARPA, the Knight Foundation, the Democracy Fund, and Indiana University internal support across various OSoMe projects. Specific Botometer maintenance funding not publicly broken out. - **Business model:** Free public research tool. No commercial product. A separate listing on RapidAPI (Botometer Pro) provides paid programmatic access for higher-volume users; the public web interface and Python client remain free. - **Open source:** yes - **Built for journalism:** no **Known issues:** Original Botometer disabled in 2023 after X ended free researcher API access. Replacement Botometer X operates in archival mode only — no scores for accounts created or active after May 31, 2023, and existing scores are based on pre-June 2023 data. No equivalent replacement has emerged for live X bot detection. Bot detection in general is probabilistic; high scores are signals to investigate further, not proof of automation. No support for non-Twitter platforms. --- Canonical HTML: https://fieldwork.news/tools/botometer Full dataset: https://fieldwork.news/llms-full.txt Methodology: https://fieldwork.news/methodology