The recent launch of OpenAI’s GPT-5 model has drawn significant criticism, as users report disappointing performance and lackluster responses. Despite claims of minor upgrades, many found GPT-5’s answers too brief and lacking personality, while still making basic errors. This led to fan demands for the reinstatement of the previous model, GPT-4o, which OpenAI initially removed. In a surprising move, the company acquiesced to user requests. Speculation suggests that GPT-5 is more of a cost-saving measure than a true advancement, aiming to reduce expenses as the AI industry grapples with financial pressures. The model features a dual structure, with a router selecting between a lightweight and a complex reasoning version, but initial execution faltered. Cost-cutting strategies, including limits on free user messages and unchanged memory capacity, highlight OpenAI’s challenge to maintain profitability amid rising competition. Users remain attached to earlier models, complicating future upgrades and developments.
Source link
