Sam Altman, CEO of OpenAI, made a statement that stirred up quite a bit of buzz during a Salesforce annual conference in San Francisco. He admitted that the world has legitimate reasons to fear artificial intelligence, but asked people to trust the companies in the space to do the right thing. It is the kind of statement that makes you raise an eyebrow, right?
In his own words, the world should trust that they will do the right thing because it is the right thing and because they feel the magnitude of it. At the same time, Altman acknowledged that people’s fears are justified, since the capabilities of AI tools have advanced incredibly fast. According to him, it no longer takes as much imagination as it used to in order to picture how all of this could go wrong, and that is why the world is right to be concerned.
The statement came at a pretty sensitive moment: just days earlier, a researcher who left Anthropic published a post that went viral globally, claiming that AI could wipe out all of humanity by the end of this decade if it is not brought under control. The debate around AI safety has never been this heated. On one side, industry leaders argue that the companies themselves are capable of self-regulating. On the other, voices from within the industry warn that leaving everything in the hands of those building the technology is far too risky of a bet. So what exactly did Altman say, who backed him up, who pushed back, and what is actually being built in practice to make sure this technology does not spiral out of control? Let’s break it down. 👇
What Sam Altman said at the Salesforce conference
During the event in San Francisco, Sam Altman got straight to the point: yes, artificial intelligence poses a real risk, and people are not wrong for being worried. But the solution, according to him, does not lie in government regulation. Altman argued that tech companies, especially those on the front lines of AI development, are the best equipped to understand the risks and make the right decisions.
He showed a lot of confidence in that vision. In his words, the industry is going to get it right, and he was very confident in his own company’s and the broader industry’s ability to do it safely. Altman also assured that they plan to keep alignment and safety well ahead of model capabilities, and that if at any point they could not, they would slow down or stop. That kind of promise is exactly the core of the self-regulation idea he is championing.
Receive the best innovation content in your email.
All the news, tips, trends, and resources you're looking for, delivered to your inbox.
By subscribing to the newsletter, you agree to receive communications from Método Viral. We are committed to always protecting and respecting your privacy.
This position is not new coming from Altman, but it carried a different weight in this context. OpenAI is one of the most influential organizations in the AI space today, and every word from its CEO is closely watched by investors, regulators, researchers, and the global media. When he says he does not need regulation to do the right thing, a lot of people hear that as a massive company asking to operate without outside oversight. And that is where the problem starts, because the industry’s recent track record does not exactly help build that argument.
The post that set off a red alert across the industry
Just days before Altman’s remarks, a researcher who left Anthropic published a piece that quickly went viral worldwide. In the post, he stated, with a certain unsettling seriousness, that artificial intelligence has the potential to wipe out humanity before the end of this decade if no effective control mechanisms are put in place. The claim might seem over the top at first glance, but the fact that it came from someone with deep technical experience at one of the most advanced companies in the field gave the text a weight that could not be easily dismissed.
What is interesting is that other executives and AI experts responded by agreeing with the assessment, even though they did not explain exactly how they reached that conclusion or in what way AI could actually pull off something that drastic. That lack of technical detail did not stop the backlash from growing. If anything, the alarm only brought more scrutiny to artificial intelligence development in the days that followed.
In response, Anthropic CEO Dario Amodei went so far as to call for the pace of all AI development to be slowed down and urged governments to regulate the industry. The viral post, by the way, was praised by several heavyweight figures in the space, including Altman himself, Google DeepMind cofounder Demis Hassabis, and Elon Musk, owner of the social platform X and the AI assistant Grok. In other words, even among the leaders there is acknowledgment that the risks are real.
Self-regulation or outside oversight: where is the line
The case for self-regulation made by Sam Altman is not a lone position in the industry. Shortly after his comments, Meta chief Mark Zuckerberg wrote on X that every AI lab has the ability and the incentive to take its own steps toward tools and models designed to be safe. According to him, any lab that does not focus on alignment will fall behind, and companies face significant legal liability if their models cause harm, which is already a strong incentive to prevent problems.
Jensen Huang, head of Nvidia, followed the same line during the conference. He stated that AI companies themselves should decide whether new versions of the technology should be released, without being managed by outside forces. In his view, no new laws or regulations are needed, and there should be no false choice between the speed of innovation and the safety of AI products. It is worth noting that Nvidia is currently the most valuable company in the world by market cap, driven by demand for its AI computing chips.
Huang also described safety as paramount, but framed it as an engineering problem. He went on to say that if at any point the leader of an AI company does not have confidence in their own product, they should simply choose not to ship it. In his view, it is obvious: move as fast as possible, but hit the brakes whenever you feel the organization is not in control.
On the opposite side, safety experts and independent researchers point out that the history of technology is packed with examples where self-regulation simply did not work. Jack Clark, executive and cofounder of Anthropic, said that leaving AI as a completely unregulated industry would be like rolling the dice with enormous stakes. Meanwhile, Patrick Hillman, chief operating officer at Logical Intelligence, a company led by Yann LeCun, noted just how little faith people have in tech companies to actually act in the public interest.
According to Hillman, the only institution Americans might trust less than Washington these days is Silicon Valley, and he insisted both deserve that skepticism. He issued a direct challenge to the companies: if you believe what you are building is dangerous, show what you are willing to stop doing. It is a provocation that cuts deep into the credibility of the self-regulation argument.
What OpenAI is actually doing in practice
As much as Sam Altman’s remarks drew criticism, OpenAI is not completely empty-handed when it comes to safety. Leaders at OpenAI and Anthropic have said they recently started working toward some kind of agreement among companies on safety standards. During an appearance at the conference, Dario Amodei said that Anthropic is now in dialogue with the rest of the industry about committing to better safety standards and checks in the development of AI tools.
Amodei also noted that one of the biggest surprises of the AI boom was not so much the advancement of the technology itself, but its broader impact. According to him, nobody realized it would cause companies to grow so fast and become central to so many things in such a short period of time. That rapid growth is precisely what ramps up the pressure for more concrete answers about risks and responsibilities.
OpenAI executive Chris Lehane said the company is also working with other labs to advance frontier AI standards, building a voluntary effort right now, with or without government support. Among those labs are Anthropic and Google DeepMind. Lehane summed up the philosophy behind it by saying that with stakes this high, you cannot let perfect be the enemy of good.
The central question, at the end of the day, is not whether OpenAI has good intentions. It probably does. The question is whether good intentions are enough when the systems being developed have the potential to impact billions of people. Regulation does not exist because companies are necessarily bad actors. It exists because it creates clear responsibilities, accountability processes, and mechanisms to fix mistakes when they happen. And in that sense, the debate that Sam Altman sparked with his statement, even unintentionally, might be more valuable than the statement itself. 🤔
