Less than 24 hours after OpenAI’s much-anticipated GPT-5 launch, early feedback from users suggests the rollout may not have met the company’s high expectations.
Launch and Initial Reactions
During a one-hour livestream, OpenAI CEO Sam Altman and his team demonstrated GPT-5’s improved reasoning, multi-step processing, and enhanced personalization features. Positioned as a “PhD-level” upgrade over the GPT-4o series, the model was presented as a significant leap forward in AI capability.
However, within hours of the reveal, social media platforms, particularly Reddit, were filled with critical reactions. A thread titled “GPT-5 is horrible” quickly gained nearly 3,000 upvotes and over 1,200 comments, with many users expressing dissatisfaction and asking for older models like GPT-4 to return.
Poor Responses, Slow Output, and More
Common concerns include perceived declines in response quality, slower output, and limitations for paid subscribers. GPT-5’s new “Thinking” mode, intended for complex reasoning, is capped at 200 messages per week for ChatGPT Plus users. Additionally, Plus subscribers now have fewer model choices as OpenAI has consolidated offerings under GPT-5, claiming it can self-select reasoning depth as needed.
Some users likened the launch to “shrinkflation,” arguing that while GPT-5 scores higher on benchmarks, its practical usability feels diminished. Others voiced frustration that the previous, more reliable models were removed rather than offered alongside the upgrade.
Expectations Too High?
The criticism follows a marketing build-up in which Altman teased GPT-5’s debut with a dramatic Star Wars-inspired post, hinting at a game-changing innovation. While GPT-5 shows measurable performance improvements in testing, many see the update as incremental rather than revolutionary, especially compared to the impact of ChatGPT’s initial release in 2022.