San Francisco, widely known as a global hub for technology and innovation, recently witnessed protests from AI safety groups demanding a pause on frontier AI development to mitigate potential existential risks. Nearly 200 protesters marched under the banner “Stop the AI Race.” The demonstration began outside the headquarters of Anthropic before moving to the offices of OpenAI and xAI.
According to the event sign-up page, the group demanded that CEOs of major AI companies publicly commit to pausing AI development provided every other AI lab agrees to do the same. The protesters came from different organisations such as Pause AI and QuitGPT. The participants included academics and former AI lab employees.
Organisers argued that AI labs are violating safety promises and increasing the risk of human extinction by accelerating AI development. As per the event page, the protest was triggered by claims that Anthropic dropped its commitment in February to pause development if its AI became too dangerous. It also pointed to OpenAI weakening its safety commitments as it restructures into a for-profit corporation.
However, leading AI companies including Anthropic, OpenAI, and xAI did not issue public statements directly responding to the weekend demonstrations.
Experts raise concerns over generative AI
Ever since OpenAI introduced ChatGPT to the general public in 2022, several experts have raised concerns about the technology’s growing capabilities. Among them is Geoffrey Hinton, often referred to as the “Godfather of AI.”
He previously highlighted three kinds of harm that artificial intelligence could cause.
“One is bad actors using it to do bad things like cybercrime, corrupting elections, or launching nasty autonomous weapons. Another is causing massive loss of jobs — the large companies aren’t thinking about what happens when they replace most workers with AI. The third thing is AI itself taking over because it’s a better form of intelligence,” Hinton said.
Some of the predictions made by Geoffrey Hinton are already beginning to surface.
In India, cyber scams have become a major concern for the government and its agencies. Scammers are increasingly using advanced AI technologies for convincing phishing attacks, voice and video deepfakes, and automated malware.
There have also been instances in the recent US-Iran war where AI tools have reportedly been used to process massive amounts of data—including intercepts and satellite imagery—to identify and prioritise targets such as military infrastructure and leadership sites.
Governments working on AI regulations
At the same time, governments worldwide are actively working to develop artificial intelligence regulations and governance frameworks. These efforts aim to balance the need for innovation with risks related to safety, privacy, and bias.
In India, the India AI Governance Guidelines provide a framework that seeks to balance AI innovation with accountability, and technological progress with safety.
Meanwhile, the European Union has enacted the world’s first comprehensive AI law, which entered into force in August 2024, with major provisions becoming enforceable by August 2026.
Balancing AI innovation and risks
Artificial intelligence holds immense potential, but the risks associated with it are also real. As the technology continues to evolve, companies are trying to mitigate these risks. For instance, xAI recently paused few features of Grok chatbot after it created non-consensual explicit images of women. Similarly, Anthropic recently locked horns with the US government when it denied the use of its chatbot for surveillance on US citizens.
Despite these efforts, many experts believe there is still a long way to go in ensuring AI systems remain safe and responsible.
