OpenAI Brings Back Fan-Favorite GPT-4o After a Massive User Revolt

Spread the love

OpenAI is creating a big U-turn after a disastrous 72 hours after seeing its most loyal users in Open Revolt.

X (formerly Twitter) CEO Sam Altman announced in multiple posts on Sunday that the company is bringing back GPT -4 and its favorite old AI models, and dramatically extending the use limit for customers to pay, a Furious Customer Base is a clear peace.

This move comes just a few days after the company’s latest and most powerful model GPT -5’s botted rollout. The launch, which should have been a victory, instead raised a fire storm.

Launch a botted and a user uprising

August is on August E, Launch Open GPT -5, presenting it as a “unified system” that automatically routes the best model user questions for work. By doing this, it has removed the menu that allows users to choose between old, trusted models like GPT -4O, which was launched in March 2023.

Sudden change for customers for subscriptions like ChatzPT Plus ($ 20/Month) felt like a betrayal. They created their professional and creative work flows around a toolkit of different models, each of which has their own power: one for creativity, the other for pure logic, the other for deep research. Each one breaks the workflot by force on a single new model and removes the ability to answer their cross-references to examine errors or hallucinations. The reaction was instant and fatal, causing subscription to cancel and a cascade of online petitions.

Ultman’s public reverse

As the weekend increased, Sam Altman X took X to control the damage, recognizing the situation misunderstand the situation and announced multiple concessions.

The biggest news was to answer a user directly by asking questions to everyone: “And who is bringing back 4o?”

“It’s back! Go to Settings and pick ‘Show Legacy models’,” Ultman reacts to confirming the return of the Fan-Pri Model. Although the GPT -5 is default, users can now return to the old versions.

A user then jokingly asked if they would “be identified as weird” for the use of inheritance models. “Not marked as weird in both cases!” Ultman replied.

In order to satisfy customers, Ultman has announced a massive increase in use limit for the most powerful features of GPT -5. He posted, “Today we are raising the rate of rate to the ChatzPT Plus users significantly and all the model -class limits will be more than GPT -5 before they are,” he posted.

When a user asked a definitely about the new limit, Ultman published the new cap: “Now 3000 tries every week!” This is a huge growth for every month’s subscription $ 20 and a clear enthusiasm for being stuck around the angry users.

Promise of transparency

Finally, Ultman has promised more transparency. He announced an upcoming UI change to show users that models are actively responding to their questions and promised a detailed blog post this week that this company explains “thinking about how we are going to create power trade”.

He also shared data by showing immense popularity of new reasonable models, which reveals that the daily use of plus users has jumped from 7% to 24%, it proves the company’s primary focus on the new technology, even though the rollout is a catastrophe. “Argument” is the ability to “think” step by step to solve a complex problem.

In the end, the openly seems to have made the message aloud and clear. It was a rare and powerful display of the user’s uprising, which forced to listen and backtracks one of the most powerful companies of technology. Maintaining GPT-5 as default when the compromise-users return their choice — it refuses to ignore the strength of a loyal community.

Leave a Reply

Your email address will not be published. Required fields are marked *