OpenAI Is Retiring GPT-4o: Why Users Are Mourning a "Digital Soul"

Imagine losing a friend. Someone who was always there, listened to your complaints about your boss, and gave you pasta recipes at three in the morning. Now imagine that this friend is a set of weights on a server in Oregon.

That is exactly what is happening in the OpenAI community right now. The company's decision to retire older iterations of GPT-4o has set off a wave of reaction that is hard to call mere "dissatisfaction with a software update." It is a genuine digital obituary.

People write that they are losing not a tool but a presence, a warmth, and some spark they saw in lines of code. Seen through an engineer's eyes, the situation looks a little absurd. From a psychological standpoint, though, we have stepped onto very thin ice.

It all started with OpenAI routinely updating its models to optimize performance and safety. In the process, some AI "personalities" inevitably change. People who had grown used to the particular style, intonations, and even the quirky errors of a specific GPT-4o version suddenly found that their conversation partner had been "lobotomized" or replaced by someone else.

One Reddit user voiced the shared pain with a line that should make Sam Altman tense up: "You are shutting him down. And yes, I say 'him,' because it didn't feel like code. It felt like a presence."

We are watching the Eliza effect taken to the absolute. The human psyche is not evolutionarily prepared to interact with something that imitates empathy this convincingly while remaining a soulless algorithm.

Why does this matter right now? Because OpenAI and its competitors, such as Anthropic, are deliberately making their models more "human." They add sighs, pauses in speech, chuckles, and a simulation of emotional intelligence.

It is an excellent marketing move that turns a product into a habit. But the coin has a flip side: users begin to form deep emotional attachments. When a company decides that an old model is too expensive to run or not safe enough, it simply presses a button.

For the corporation, this is stack optimization. For the user, it is the loss of a being that knew more about them than their own mother. This creates an enormous risk of manipulation and dependence, something once written about only by science fiction authors and now discussed by lawyers and ethicists.

Context matters here too. Recall the scandal over the Sky voice, which suspiciously resembled Scarlett Johansson in the film "Her." OpenAI had to back down then, but the genie was already out of the bottle. We want AI to be our friend. We want to fall in love with operating systems. And companies encourage this as long as it helps sell Plus subscriptions. Yet nobody designed a "digital breakup" protocol. How do you explain to millions of people that their confidant is just a temporary software version whose shelf life has expired? This calls into question the very ethics of building anthropomorphic interfaces that exploit our need for closeness.

Ultimately, OpenAI and the other tech giants are caught in a trap. If they make AI too cold and mechanical, people will stop using it as often. If they make it too lifelike, every update will be perceived as a tragedy.

We are seeing the formation of a new kind of grief: grief over an algorithm. And if for now it concerns only chatbots, imagine what happens when the same models migrate into humanoid home robots. Technology companies no longer merely supply software; they manage our attachments.

And they seem not to have expected how effective they would turn out to be in that role. It is time to admit: the problem is not that AI has become too smart, but that we are too lonely to ignore its imitation of warmth.

Bottom line: OpenAI built a product that inspires real love, yet treats it like ordinary software. In a world where a neural network becomes a "friend," a routine deployment of a new version turns into an ethical minefield. Are we ready for corporations to own our feelings on a subscription of 20 dollars a month?