AI is the new world, and people are already obsessed with its wonderful outputs, whether it is creating images for self-happiness or to get their daily corporate job done. But one forgets to notice the lives of humans behind the AI. The people who are executing the functions and making it efficient to excel in its performance day by day.
Many such scientists and engineers working in AI companies have spoken out recently about their fear of the technology getting far better every day, citing it as a threat to human intelligence. Let’s take a look at some examples of such recent warnings given by those working inside these offices.
Jacob Coxon was a former OpenAI and Anthropic researcher who has recently spoken about the AI threat in today’s world. After working at Anthropic for three years as a researcher, he decided to call it quits, but not silently, instead sparking a wide conversation about the upcoming threats of the intelligence. He posted on his X account, stating,
"I resigned from Anthropic today. I spent the last three years doing pretraining research at both OpenAI and Anthropic. Neither company is acting responsibly. They are racing straight to self-improving superintelligence and gambling with our lives."
I resigned from Anthropic today. I spent the last three years doing pretraining research at both OpenAI and Anthropic. Neither company is acting responsibly. They are racing straight to self-improving superintelligence and gambling with our lives. More thoughts below.
He did not stop there. In a series of threads, he also alleged that the machines will eventually kill humans and can possess the power to hack anything, which worries the seniors of these firms.
Coxon also urged his colleagues to take necessary steps to counter the effect and to stop moving forward with technology that could go out of hand.
Anthropic’s Alignment Science Lead, Evan Hubinger, agreed with the points raised by Coxon, personally believing the chances of it happening in the next decade are above 10 percent.
“Jacob is correct here—we really do earnestly believe AI could kill all humans! I personally think it is >10% within the next decade. I believe Anthropic is trying its best, but we do not yet have a plan to solve alignment for superintelligence and are not clearly on track to.” he posted on X.
Jacob is correct here—we really do earnestly believe AI could kill all humans! I personally think it is >10% within the next decade. I believe Anthropic is trying its best, but we do not yet have a plan to solve alignment for superintelligence and are not clearly on track to. https://t.co/QAIHiFP3QZ
In February 2026, Mrinank Sharma also left Anthropic, citing the ‘world is in peril’ warning. In the resignation letter he shared on X, Sharma stated that he is leaving the AI company because of his concerns over the consequences of the continued rapid rise in machine intelligence.
As for his next plans, Sharma stated that he wants to venture into poetry and community building, ending the letter with his favourite lines from a poem.
“I hope to explore a poetry degree and devote myself to the practice of courageous speech,” he added.
Today is my last day at Anthropic. I resigned.Here is the letter I shared with my colleagues, explaining my decision. pic.twitter.com/Qe4QyAFmxL
OpenAI chief scientist Jakub Pachocki also voiced his concern over the rising AI intelligence and its consequences. In his recent blog, ‘An Alien Mind’, he stated,
“I am concerned no one is prepared for the consequences of a continued rapid rise in machine intelligence.”
(- Jakub Pachocki, OpenAI scientist)His post came shortly after OpenAI introduced GPT-6 Astra, which the company described as its most intelligent and aligned model.
OpenAI said in July that one of its autonomous AI systems had carried out a hack of technology platform Hugging Face, describing the episode as an unprecedented incident involving AI agents acting with limited human intervention.
Moreover, Pachocki also ensured that the company is working on ‘defensive’ measures to tackle the issue and to align the machine interests with those of humans.
Jan Leike is another person who called it quits when ChatGPT launched its model GPT-40.
He was a safety researcher and co-head of superalignment at OpenAI who called it quits with the AI company after certain disagreements within the organisation. His main reason, which he stated for his resignation, was the lack of safety culture in the company, and leading to a break-even point. He followed to call it quits after his other colleagues, Ilya Sutskever and John Schulman, also decided to leave the firm.
Responding to Leike’s allegations on X, OpenAI CEO Sam Altman thanked Leike for his contributions and ensured that the company will work towards all necessary changes.
Yesterday was my last day as head of alignment, superalignment lead, and executive @OpenAI.
The fact that researchers within leading AI companies openly acknowledge the possibility of catastrophic outcomes adds to public anxiety about the technology's rapid development.
In terms of legal intervention, the recently launched European Union’s AI Act is among the few measures requiring companies to prove their new AI models. But the limitations of such rules in a few countries will not make much difference when compared to a global threat.
Even MIT professor Joseph Weizenbaum, who created the world’s first chatbot, ‘ELIZA’ in 1966, has spoken about his concerns at the time when the dependency on his invention was adopted by humans so quickly, warning that it was a threat to humankind.
To conclude, one can only hope that boons like AI intelligence could stay in its positive realm rather than becoming a bane to its own creators.