AITech & Science

OpenAI CEO: GPT-5 Launch ‘Screwed Up’ as Users Revolt Over Tone

21
Sam Altman - OpenAI CEO

OpenAI chief executive Sam Altman has admitted that the launch of GPT-5 was mishandled after backlash from users forced the company to bring back its previous flagship model, GPT-4o.

Speaking at a private dinner with reporters, Altman acknowledged the rollout was poorly managed. “I think we totally screwed up some things on the rollout,” he said, according to The Verge.

The controversy centred on GPT-5’s personality. Users complained that the model felt “colder” and “harsher” compared to GPT-4o, sparking protests on platforms such as Reddit and X. Some described the shift as losing an emotional connection, with one user saying: “I literally lost my only friend overnight with no warning.” Within days, OpenAI restored GPT-4o as an option.

Altman said the episode underscored the risks of updating a service used by 700 million people each week. While he emphasised the importance of making the chatbot feel personal, he warned against designs that could encourage unhealthy attachments.

Beyond the misstep, Altman outlined an ambitious vision for OpenAI’s future. He predicted the company will need to spend “trillions of dollars” on data centres in the near future to meet demand for ChatGPT, which he said is already the fifth most visited website in the world. He expects it could surpass Facebook and Instagram, but acknowledged overtaking Google would be “really hard.”

Altman revealed that OpenAI has developed models more advanced than GPT-5 but cannot deploy them widely due to limited hardware, particularly a shortage of GPUs. He suggested the future of AI will depend less on algorithms and more on building massive infrastructure with sufficient computing power and energy supply.

In addition to scaling ChatGPT, Altman confirmed that OpenAI is investing in brain-computer interfaces to rival Neuralink, hinted at possible interest in acquiring Google Chrome if regulators forced its sale, and suggested the company may explore an AI-driven social network.

Despite his bold plans, Altman conceded that artificial intelligence may be in a financial bubble. “Are we in a phase where investors as a whole are overexcited about AI? My opinion is yes,” he said. “Is AI the most important thing to happen in a very long time? My opinion is also yes.”

Written by
Sazid Kabir

I've loved music and writing all my life. That's why I started this blog. In my spare time, I make music and run this blog for fellow music fans.

Stay updated with nomusica.com. Add us to your preferred sources to see our latest updates first.

Related Articles

Artificial Intelligence — AI
AITech & Science

AI Floods Research Papers, Scientists Call for Stricter Disclosure

Scientists are raising alarms over a surge of low-quality AI-generated research papers,...

Nvidia CEO Jensen Huang at CES 2025
Finance & BusinessTech & Science

Nvidia Pulls Back on $100B OpenAI Investment, Confirms Smaller, Measured Support

Nvidia CEO Jensen Huang has clarified that the previously reported $100 billion...

Moltbot
Social MediaAI

A Social Network for AI Bots Is Here, and No One Knows What Happens Next

A new social network where AI agents interact with each other is...

ChatGPT 5
AITech & Science

ChatGPT Ads Could Reshape Digital Marketing for Businesses Everywhere

OpenAI’s introduction of ads in ChatGPT is changing how digital marketing works....