On February 26, 2024, the trending topic on social media was the backlash against Google’s Gemini AI model. Memes, screenshots, and links mocking or criticizing the inaccuracies of Gemini flooded feeds throughout the weekend.
One particular incident involved X (formerly Twitter) users sharing screenshots of Gemini output that compared the societal impact of Elon Musk’s memes to that of Hitler. VC Marc Andreessen even joined in on the fun by reposting offensive outputs and claiming they were intentionally programmed with hate by Gemini creators.
This negative response marks a stark contrast to the initial excitement surrounding Gemini’s release in December, where Google was lauded for its foray into advanced AI models that could rival GPT-4. Just over a year ago, Google faced a similar crisis with ChatGPT, prompting CEO Sundar Pichai to overhaul the company’s AI strategy to stay competitive.
Google’s hesitation to release sophisticated language models like Gemini was fueled by the fear of backlash, as seen in previous instances such as the LaMDA debacle of 2022. In that case, Google engineer Blake Lemoine believed the AI model he worked on was sentient, leading to a controversial investigation into its discriminatory language.
With Google’s history of AI mishaps and the fast-paced nature of technological advancements, the backlash against Gemini was almost inevitable. These incidents highlight the delicate balance Google must strike between innovation and user satisfaction in the AI landscape.
The AI Impact Tour – NYC
Microsoft and VentureBeat are joining forces for The AI Impact Tour in New York on February 29. Join us for an exclusive discussion on navigating the risks and rewards of AI applications. Request your invite today!