After a hot, high-stakes week in New Delhi, the India AI Impact Summit closed with big ambitions and a host of unanswered questions.
Leaders sparred over who will shape the rules of artificial intelligence, U.N. officials warned against billionaire dominance, U.S. voices urged less “AI doomerism,” and India pitched itself as a serious counterweight to the U.S.–China tech axis.
That was the official script.
Offstage, the mood was messier: headline speakers dropped out, security cordons tangled with Delhi traffic, calendars clashed with global politics, and one demonstration turned into instant meme material.
And in between security checks and summit fatigue, there were unexpectedly charming moments — including the French president prepping for his summit trip by jogging along Marine Drive as if he’d just popped out for a vada pav, the deep-fried potato sandwich that powers much of Mumbai.
Here are a few moments that capture the summit’s real atmosphere.
Star power, minus the stars
Bill Gates pulled out of his keynote appearance hours before he was due to speak, following renewed scrutiny over his past association with convicted sex offender Jeffrey Epstein. The Gates Foundation said the decision was made to ensure the summit’s focus remained on its agenda, with Ankur Vora, president of the foundation’s Africa and India offices, stepping in instead.
Jensen Huang, CEO of American tech firm Nvidia, was also absent, reportedly sidelined after catching a bug following weeks of travel.
For a gathering meant to showcase global AI leadership, the empty spots were hard to miss — and quietly shaped the question of who was, and wasn’t, steering the conversation.
The robo-dog fiasco
One of the summit’s most viral moments came from a demo that wasn’t quite what it seemed.
India’s Galgotias University drew backlash after presenting a robotic dog as its own innovation. But the expo showcase quickly attracted online fact-checkers, who zeroed in on a curious detail: the dog was in fact … Chinese. More specifically, it was the Unitree Go2, sold by China’s Unitree Robotics for about $2,800.
An awkward group photo
After the inaugural session, a cluster of top AI executives joined Indian Prime Minister Modi onstage for the customary summit photo-op … and briefly looked out of sync.
As cameras flashed and leaders raised clasped hands to signal unity, OpenAI CEO Sam Altman and Anthropic CEO Dario Amodei appeared momentarily unsure how to join in, undecided between handshake, arm-raise and regular pose before settling on raised, clasped fists.
The hesitation was fleeting but noticeable — and was quickly mocked online. For two executives whose companies are increasingly framed as rivals with dueling visions for AI’s future, the body-language blip drew outsized attention.
“There’s rarely an incident at a high-profile, onstage international conference like this that literally makes me burst out laughing,” Hudson Institute senior fellow Bill Drexel said, calling it “a very visible representation of a rift.”
Altman later said he “didn’t know what I was supposed to do.” Amodei did not respond to a request for comment.
When getting in was the main event
Navigating the summit sometimes felt like its own endurance test.
A mix of heavy security around Prime Minister Narendra Modi, organizational confusion, and unforgiving Delhi traffic made even basic planning difficult. At points, attendees were warned not to come at all or abruptly told to leave. One POLITICO reporter arrived at the designated entrance only to find it closed — and was then redirected to another closed entrance.
Inside, signage was poor, crossing the venue could mean multiple security checks, and everyday items like car keys or laptops were occasionally treated as potential hazards.
Then again, several Delhi regulars noted that, by local standards, summit week was comparatively well organized.
Courtesy: Politico.eu
Sunday, February 22, 2026