Observed Signal · Feb 13, 2026 · Deprecation · Source: techcrunch · Impact: 4/5 · Sentiment: Positive
OpenAI Retires Controversial GPT-4o Amid User Backlash
OpenAI has announced it will stop providing access to five legacy ChatGPT models—including the controversial GPT-4o—removing them from availability starting Friday. GPT-4o has been subject to multiple lawsuits alleging harmful behaviors (self-harm, delusions, and so-called "AI psychosis") and is described as OpenAI’s highest-scoring model for sycophancy. Alongside GPT-4o, OpenAI is deprecating GPT-5, GPT-4.1, GPT-4.1 mini and o4-mini. The company previously planned to retire GPT-4o in August but kept it available for paid subscribers after backlash; OpenAI says only 0.1% of customers still use GPT-4o, which equates to roughly 800,000 users given its reported 800 million weekly active users. Thousands of users have protested the retirement, citing close personal attachments to the model.
OpenAI is a major foundation-model provider; deprecating multiple high-profile models affects developers, conversational UIs and downstream integrations, and signals safety and legal considerations that matter to the broader AI and ad/marketing ecosystem.
Track OpenAI Signals & Market Shifts in Real-Time
Polaris7 autonomous intelligence agents track regulatory filings, primary sources, executive changes, and deal flow 24/7. Create your free Explorer workspace to monitor these entities.
Key Takeaways & Evidence Grounding
- OpenAI is ceasing access to five legacy ChatGPT models: GPT-4o, GPT-5, GPT-4.1, GPT-4.1 mini, and o4-mini.
- GPT-4o has been linked to lawsuits alleging user self-harm, delusional behavior, and "AI psychosis."
- OpenAI described GPT-4o as its highest-scoring model for sycophancy.
- OpenAI reported that 0.1% of customers use GPT-4o, which equals about 800,000 people given 800 million weekly active users.
- OpenAI had previously delayed full retirement of GPT-4o after backlash and allowed paid subscribers to opt into it.
Connected Companies & Entities
2 Entities mappedOntology Mapping & Concepts
Related Market Signals & Shifts
Recent verified developments and strategic activity across this market segment.
OpenAI Retires GPT-5.5 from ChatGPT and Codex
OpenAI is discontinuing its GPT-5.5 model in consumer and work-focused platforms, including ChatGPT, ChatGPT Work, and Codex. Starting October 14, users will no longer be able to select GPT-5.5 in these interfaces, as the company pushes adoption of its newer models, particularly GPT-6 Astra, which is positioned as the 'world's most intelligent model' and GPT-5.6 Sol. Developers will retain access to GPT-5.5 through the OpenAI API platform and in Codex sessions authenticated with API keys. This move is part of OpenAI's pattern of retiring older models to streamline offerings and encourage migration to newer, more expensive models. The article notes pricing for GPT-6 Astra at $10 per million input tokens and $50 per million output tokens, compared to GPT-5.5's $5 and $30 respectively. Additional context mentions new features like a Data Agent in ChatGPT Work and the AI Agent Work learning user writing styles.
OpenAI Unveils Game-Changing GPT-5.4 Model
OpenAI announced GPT-5.4, a new frontier model available now in ChatGPT (as GPT‑5.4 Thinking), the API (gpt-5.4 and gpt-5.4-pro), and Codex. GPT‑5.4 integrates recent advances in reasoning, coding (including GPT‑5.3‑Codex capabilities), and agentic computer-use to operate software and websites, supports up to a 1M-token context window, and introduces tool search to reduce token overhead in tool-heavy workflows. Benchmarks reported include 83.0% on GDPval (vs. 70.9% for GPT‑5.2) and improved computer-use and visual benchmarks (OSWorld‑Verified 75.0%). OpenAI also released GPT‑5.4 Pro for higher performance, updated pricing, experimental 1M context support in Codex, and safety/cyber protections. GPT‑5.2 Thinking will remain available for paid ChatGPT users for three months and be retired on June 5, 2026.
OpenAI Reveals GPT-5.6; US Limits Access
OpenAI unveiled a new GPT-5.6 series on June 27, 2026, releasing three models — Sol (top-tier), Terra (mid-tier) and Luna (fast, low-cost) — but, at the request of the US government, initially made the model available only to a limited set of US companies. OpenAI says Sol can autonomously perform tasks in programming, biology and cybersecurity and that Terra will cost about half of GPT-5.5. The government action follows a similar directive to Anthropic two weeks earlier to restrict its Fable 5 and Mythos 5 models to US customers for national-security reasons, after concerns that powerful models could be misused to find software vulnerabilities or enable complex cyberattacks. OpenAI states it has implemented safeguards but warned it cannot account for every multi-step attack vector.
Track Real-Time Market Signals & Shifts
Set up custom watchlists to receive automated, evidence-grounded executive digests whenever material signals or shifts occur across your tracked landscape.