DeepMind CEO Questions OpenAI’s Ads in ChatGPT
Google DeepMind CEO Demis Hassabis said he is “surprised” that OpenAI has moved to include ads in its ChatGPT chatbot. In an interview with Axios at Davos, Hassabis said Google is taking a more cautious approach to monetizing AI services.
DeepMind’s Approach
Hassabis emphasized that his team does not feel pressure from Google to rush into ads, despite their importance to Google’s core business. He described DeepMind’s process as careful and methodical, focusing on trust and user experience.
OpenAI’s Ads Move
OpenAI recently announced it will test ads for non-subscribing users of ChatGPT, which has around 800 million weekly active users. Hassabis said ads can work if done well, but raised concerns about integrating them into a digital assistant meant to be helpful.
User Experience Concerns
Hassabis highlighted that chatbots differ from search engines. Search ads are targeted based on intent, while chatbots aim to act as personal assistants. Ads in conversations could undermine trust and reduce usefulness.
Google’s Focus on Personalization
DeepMind is enhancing Google’s Gemini AI with personalization features. Users can allow Gemini to reference Gmail, Photos, Search, and YouTube to provide tailored responses. Hassabis said DeepMind wants to build functionality first and explore ads thoughtfully later.
Lessons from the Market
Hassabis cited past pushback against OpenAI’s app suggestions and Amazon Alexa’s ad attempts. He stressed that users expect AI assistants to help them, not sell to them.
“We don’t feel pressure to make knee-jerk decisions,” he said. “We take a scientific and rigorous approach to every step, whether technology or product.”
