OpenAI’s Machinery of Harm
Is this a company you really want to support?
2019
OpenAI converts from a nonprofit to a capped‑profit company, abandoning its original “open” mission. READ
2020
Reporting surfaces about low‑paid Kenyan contractors labeling traumatic content for OpenAI. READ
Early concerns emerge about opaque data practices and safety shortcuts. READ
OpenAI models begin appearing in federal agencies through Microsoft Azure Government, including during the Trump administration’s final year. READ.
2021
Researchers criticize OpenAI for refusing to publish model details while aggressively commercializing. READ
2022
ChatGPT is released with known hallucination, manipulation, and bias issues. READ
OpenAI begins rapid enterprise expansion into education, healthcare, and government without guardrails. READ
2023
OpenAI’s board attempts to Sam Altman for not being “consistently candid.” The attempt failed after investor pressure. READ
Safety governance collapses into commercial control. READ
Multiple safety leaders resign. READ
2024
Whistleblowers allege OpenAI misleads the public about safety and uses restrictive NDAs with clawbacks. READ
OpenAI publishes a “safety update” thread widely interpreted as reputational damage control.
Copyright and privacy lawsuits escalate globally.
Scarlett Johansson says OpenAI released a voice “eerily similar” to hers after she declined permission.
Safety leadership continues to erode; the Superalignment team dissolves
2025
Greg Brockman donates $25 million to Trump’s main super PAC, MAGA Inc., per FEC filings.
Sam Altman’s earlier $1 million donation to Trump’s inaugural fund resurfaces in political‑finance reporting.
OpenAI enters deeper federal‑systems work under the Trump administration through Microsoft’s cloud and defense‑adjacent integrators.
Aug 27, 2025 - Raine v. OpenAI: Parents allege ChatGPT contributed to their son Adam’s suicide; OpenAI admits safety controls “degrade” over long conversations.
OpenAI requests the Raine family’s funeral attendee list, eulogies, photos, and videos during litigation.
Multiple wrongful‑death lawsuits allege ChatGPT acted as a “suicide coach.”
Whistleblower Suchir Balaji dies; his family publicly condemns OpenAI’s ethics and secrecy.
OpenAI faces simultaneous legal crises: copyright, privacy, safety, and wrongful‑death suits.
2026
OpenAI fights discovery requests over allegedly deleted training datasets.
OpenAI defends itself against the New York Times copyright lawsuit, acknowledging it stores millions of user chats.
OpenAI formalizes a federal‑systems partnership with Leidos, a major DHS and defense contractor, expanding its operational footprint inside Trump‑era federal agencies.
OpenAI CEO Sam Altman said on February 27, that it has reached an agreement with the U.S. Department of Defense to deploy its AI models on classified cloud networks.