Who is Mrinank Sharma? The Tech Savy’s Resignation From Anthropic AI In Spotlight
The Indian-born artificial intelligence researcher Mrinank Sharma is recognized as one of the leading experts in AI safety. He held the position of Safeguards Research Team leader at Anthropic AI until February 9, 2026. Sharma holds a DPhil in Machine Learning from Oxford and a Master of Engineering from Cambridge, which he used to solve real-world AI challenges. His research focused on understanding how large language models create security risks, which he addressed by developing AI protection systems and methods to prevent AI-assisted bioterrorism, effectively acting as a protective barrier against dangerous AI activities.
Sharma made an unexpected decision to leave Anthropic AI, citing that “the world is in peril,” referring not only to AI threats but also to multiple global crises. He departed to explore significant questions through his poetic, philosophical, and scientific work, demonstrating that even top AI experts need time to reflect on their core values.
You Might Be Interested In
What Was Mrinank Sharma’s Role at Anthropic AI?
Mrinank Sharma worked as the Safeguards Research Team leader at Anthropic AI, where he researched methods to protect artificial intelligence systems and develop safer ways to use large language models such as Claude.
He worked to improve jailbreak security weaknesses through his development of automated testing methods and his research on AI-based bioterrorism defense strategies. Sharma’s research targeted key vulnerabilities that he identified as “many-shot jailbreaking,” which he addressed through his development of “Constitutional Classifiers” that filter out dangerous material.
The team created advanced software tools and security protocols under his management, which helped organizations protect sensitive situations from AI abuse, making Anthropic AI an international standard for artificial intelligence security and responsible technology development in the fast-changing AI industry.
Why Did Mrinank Sharma Resign From Anthropic AI ?
In early February 2026, Sharma stepped down, sounding the alarm on what he called a world “in peril.” He warned about multiple interconnected crises happening simultaneously, from AI and bioweapons to other emerging threats. He explained that human understanding must catch up with our technological abilities before major repercussions occur. Sharma highlighted how employees at tech companies often face internal battles, as organizations prioritize rapid progress over ethics, forcing staff into difficult moral decisions.
His resignation served two purposes: it acted as a public warning and reflected his personal values guiding his career choices. Recently, Anthropic AI itself made headlines, crashing the internet spotlight with its high-profile developments.
His Resignation Letter
In a detailed post on X, Mrinank Sharma reflected on his time at Anthropic, noting that he had accomplished key goals, including building defenses against AI risks and studying AI sycophancy. He highlighted his final project, which examined how AI assistants could subtly distort human behavior. Sharma explained that he now wants to explore questions that “have no right to go away,” invoking Zen wisdom with the phrase: “not knowing is most intimate.”
Mrinank Sharma’s Career Pivot, Creative Pursuits, And Ethical Stand
After resigning from Anthropic AI, Mrinank Sharma surprised many by announcing he won’t jump to another tech firm. Instead, he plans to pursue a degree in poetry and focus on “courageous speech,” creating a space where scientific and poetic truth intersect in technology development.
Sharma also intends to deepen his skills in facilitation, coaching, community building, and group work, blending reflection with action. Beyond AI, he is an avid poet-author of We Live and Die a Thousand Times-as well as a DJ and organizer for a spiritual “Sangha” in the Bay Area. Recently, he revealed plans to move back to the United Kingdom and “let himself become invisible for a period of time,” emphasizing introspection. His resignation also highlighted systemic pressures in tech companies that push employees to compromise values for speed and growth, sparking wider discussions on ethical AI development and corporate responsibility.
Aishwarya is a journalism graduate with over three years of experience thriving in the buzzing corporate media world. She’s got a knack for decoding business news, tracking the twists and turns of the stock market, covering the masala of the entertainment world, and sometimes her stories come with just the right sprinkle of political commentary. She has worked with several organizations, interned at ZEE and gained professional skills at TV9 and News24, And now is learning and writing at NewsX, she’s no stranger to the newsroom hustle. Her storytelling style is fast-paced, creative, and perfectly tailored to connect with both the platform and its audience. Moto: Approaching every story from the reader’s point of view, backing up her insights with solid facts.
Always bold with her opinions, she also never misses the chance to weave in expert voices, keeping things balanced and insightful. In short, Aishwarya brings a fresh, sharp, and fact-driven voice to every story she touches.
