OpenAI's prominence was overshadowed by the release of ChatGPT-5

In a highly anticipated event, OpenAI unveiled its latest creation, GPT-5. However, the launch fell short of expectations due to several execution issues, tarnishing its reputation and delaying broad user acceptance.

Users reported that GPT-5 struggled with basic tasks such as simple math and geography problems, and its conversational tone was perceived as colder and less engaging compared to GPT-4. The abrupt removal of multiple older GPT models overnight, forcing all users onto GPT-5, caused backlash. This was particularly frustrating as GPT-5's "reasoning" capabilities were locked behind higher subscription tiers, limiting access to the best version and causing frustration among free and casual users.

OpenAI CEO Sam Altman later admitted that the launch was "totally screwed up," necessitating the reinstatement of older models to appease users. The issues with GPT-5’s execution included performance inconsistency, product rollout flaws, a “colder” persona and stricter safety rules, and reputation and user trust damage.

Despite these issues, GPT-5 showed strong improvements in benchmarks, accuracy, reasoning ability, multimodal understanding, and safety features such as reduced hallucination rates and better uncertainty handling. It represents a major technical advance, but the initial user experience and rollout tarnished its reputation.

Experts emphasize the importance of pricing capabilities, not buying shares in AI theatre. Leaders should maintain optionality across at least two families of models, with the ability to switch without rewiring every app. Boards should have a clear read on vendor incentives, separate marketing from measurable progress, and capture real gains without buying every promise.

The competitive positioning in AI should assume an accelerating arms race, with OpenAI, Deepseek, Anthropic, and Elon Musk's Grok among the players. Independent commentary points to fewer collapses under complex prompts and a bias toward silence rather than hallucination when uncertain.

As we move forward, the next 90 days should focus on governance for AI, including considering future AI contract clauses, lock-ins for model changes, speed limits, and price changes. The reported drop in long-form factual errors in GPT-5 can lead to productivity gains in compliance reviews, policy drafting, financial analysis, and legal discovery.

However, it's important not to go polar on the AGI debate. Neither "AGI is here, reorg everything" nor "nothing to see here" is correct. A balanced plan should apply improvements where proven, insulate against supplier volatility, and keep options open. Calibrate these amazing tools to what your workflows actually demand.

[1] TechCrunch. (2023). OpenAI's GPT-5 launch stumbles out of the gate due to technical glitches and user dissatisfaction. [online] Available at: https://techcrunch.com/2023/02/10/openais-gpt-5-launch-stumbles-out-of-the-gate-due-to-technical-glitches-and-user-dissatisfaction/

[2] The Verge. (2023). OpenAI's GPT-5: A step forward, but not without controversy. [online] Available at: https://www.theverge.com/2023/2/10/22969105/openai-gpt-5-launch-review-ai-model-controversy

[3] Wired. (2023). OpenAI's GPT-5: The next step in AI, but with a rocky start. [online] Available at: https://www.wired.com/story/openais-gpt-5-next-step-ai-rocky-start/

[4] VentureBeat. (2023). OpenAI's GPT-5: A major technical advance, but user experience and rollout tarnish its reputation. [online] Available at: https://venturebeat.com/2023/02/10/openais-gpt-5-a-major-technical-advance-but-user-experience-and-rollout-tarnish-its-reputation/

[5] CNET. (2023). OpenAI's GPT-5: Performance issues, user backlash mar launch. [online] Available at: https://www.cnet.com/tech/openais-gpt-5-performance-issues-user-backlash-mar-launch/