Safety researchers resign from Anthropic, warning that commercial pressures and rapid AI development pose serious dangers to human society.
A wave of high profile resignations has hit top artificial intelligence developer Anthropic, as key safety researchers step down in protest against the fast paced race to build powerful digital models.
Known globally for creating the popular Claude digital assistant, Anthropic was originally founded by former tech executives to build safe, responsible computer systems.
However, senior safety experts are now walking away from their high paying positions, raising urgent alarms that major tech companies are prioritizing commercial profits, government military contracts, and speed over public safety.
These public departures highlight deep internal conflicts within leading tech firms as researchers warn that building unchecked digital tools creates severe societal dangers.
The dramatic safety resignations gained global attention after departing research leads, including former Safeguards Research Team chief Mrinank Sharma, made public statements, with additional team members stepping down through Wednesday, September 9, 2026.
Operating out of tech centers in the United States, researchers revealed that internal ethics teams face constant pressure to loosen safety guards.
The departures come as government defense agencies and corporate investors push AI firms to speed up model releases, leaving internal safety teams unable to properly test tools for dangerous risks before public deployment.
The main reason top scientists are quitting Anthropic now is that corporate profit goals and global tech rivalries are overriding original safety promises.
When tech companies race against rivals in Silicon Valley and foreign nations, executive leadership often views strict safety checks as slow speed bumps.
Furthermore, pressure from defense departments demanding broader military applications forces ethical researchers to choose between compromising their core values or leaving their jobs.
Seeing that internal safeguards cannot keep up with rapid software releases, researchers are taking their warnings directly to the public to demand global government oversight.
Warning that fast software development creates massive risks beyond simple computer glitches, former Anthropic safeguards lead Mrinank Sharma wrote in a public resignation letter that “the world is in peril,” emphasizing that unchecked technology accelerates global crises.
Detailing the constant internal struggles between maintaining strong ethical standards and meeting aggressive business targets, Mrinank Sharma added that “I’ve seen this within myself, within the organization, where we constantly face pressures to set aside what matters most”.
Reflecting the shared frustrations of safety workers watching commercial interests take over responsible development, departing safety experts warned that tech firms “are gambling with our lives” by releasing powerful software without proper safety bounds.
By resigning and speaking out publicly, former Anthropic researchers are drawing global attention to the dangers of an unregulated technology race.
Demanding strict safety baselines ensures that artificial intelligence continues to serve humanity safely rather than threatening public security.
About the Author
Jennifer Sakmufuwo Baba is a tech analyst, senior staff, and writer covering artificial intelligence, cybersecurity , and emerging technologies at TechRegard. Based in Nigeria, she's passionate about translating complex tech developments into compelling, accessible stories for diverse audiences. Her work focuses on how technology shapes innovation across Africa and globally.
