GPT-4.5 Disappoints: OpenAI’s Latest AI Model Faces Backlash Over Cost and Performance

OpenAI’s Latest AI Model Faces Criticism Amid Underwhelming Reception

OpenAI’s recently released GPT-4.5 model, touted as the “largest and most knowledgeable model yet,” has met with a lukewarm response from the public and experts alike. Despite the company’s marketing efforts, the new model is not considered a “frontier model” and has been criticized for its high cost and perceived lack of significant improvements.

The public reaction to GPT-4.5 has been notably muted, with many critics labeling it as underwhelming. The model’s significantly higher price point compared to its predecessor, GPT-4o, has raised eyebrows in the AI community. Additionally, users have reported that the model is slower and appears to prioritize emotional intelligence over analytical tasks.

Prominent AI critic Gary Marcus did not mince words, describing GPT-4.5 as a “nothingburger.” An anonymous expert in the field went further, calling it a “lemon.” The model’s cost, which is reportedly 30 times higher than GPT-4o, has led many to question whether the improvements justify such a steep increase in price.

Critics have also pointed out that GPT-4.5 continues to struggle with hallucinations and errors. OpenAI itself has acknowledged a 37% error rate in a key test, further fueling skepticism about the model’s capabilities.

In response to the criticism, OpenAI has emphasized GPT-4.5’s ability to engage in sensitive and intuitive conversations. The company claims that the model is designed to understand implicit user expectations, resulting in more nuanced responses. However, this focus on emotional intelligence has been met with mockery from some critics, who question its practical utility in real-world applications.

Internal acknowledgments from OpenAI have added to the controversy. Mark Chen, the company’s chief research officer, admitted that the model’s advancements are tracked using “vibes-based” benchmarks. This revelation has highlighted a potential disconnect between the company’s goals and public expectations for AI development.

The release of GPT-4.5 appears to have fallen short of the high expectations set by previous models. OpenAI now faces mounting criticism over the model’s cost, performance, and focus on emotional intelligence. As the AI landscape continues to evolve, the company may need to reassess its approach to future model developments to better align with user needs and industry expectations.