When the world’s most powerful AI bosses share a stage with the Prime Minister, you expect big statements about innovation, growth, maybe even world domination by large language models. What you do not expect is a viral moment over… hand-holding. Yet that is exactly what happened at the India AI Impact Summit on Thursday. A simple group photograph has now turned into the internet’s favourite metaphor for the OpenAI versus Anthropic rivalry.
First, check out the video:
OpenAI CEO Sam Altman and Anthropic CEO Dario Amodei visibly declined to hold hands during a group photo at the India AI Impact Summit, even as other leaders on stage linked arms for the ceremonial shot pic.twitter.com/J1eGShSkiK
— Reuters (@Reuters) February 19, 2026
The Handholding That Never Happened
After delivering their addresses, top tech leaders were invited on stage for a group photograph with PM Narendra Modi. The lineup was nothing short of Silicon Valley royalty: Google CEO Sundar Pichai, OpenAI CEO Sam Altman, Anthropic CEO Dario Amodei, and Meta Chief AI Officer Alexandr Wang, among others.
PM Modi stood at the centre, holding the hands of the leaders on either side of him and encouraging them to raise their arms for the cameras. Sundar Pichai and Sam Altman obliged.
But then came the moment that launched a thousand memes.
Standing next to each other, Sam Altman and Dario Amodei did not clasp hands. Instead, both raised their fists independently. No physical connection. No symbolic unity. Just two rival CEOs, side by side, carefully avoiding any visual suggestion of AI harmony.
So… Why The Distance?
To understand the “why,” you have to rewind a few years.
Dario Amodei was once Vice-President of Research at OpenAI. In 2021, he left the organisation and went on to co-found Anthropic with his sister and several former OpenAI researchers. The departure was not framed as dramatic at the time, but philosophical differences were evident.
In a 2023 interview with Fortune magazine, Amodei explained his concerns. “You needed something in addition to just scaling the models up, which is alignment or safety. You don’t tell the models what their values are just by pouring more compute into them,” he said.
In other words, bigger models were not enough. Safety and alignment needed to be central, not secondary.
OpenAI vs Anthropic: More Than Corporate Competition
Today, OpenAI and Anthropic represent two prominent approaches to building artificial intelligence.
OpenAI is widely known for moving quickly and putting tools like ChatGPT directly into the hands of users across the globe. Anthropic, meanwhile, positions itself as more focused on safety, caution and controlled deployment, especially through its Claude chatbot.
So when Altman and Amodei stood shoulder to shoulder and chose raised fists over joined hands, the optics were impossible to ignore.
Was it awkward timing? Or was it simply the most polite way to say: we may share a stage, but we do not share a playbook. In the high-stakes world of AI, even a photo-op can turn into a philosophical statement. And sometimes, the loudest message is the hand you do not hold.
