Google’s CEO Sundar Pichai has dampened expectations for major breakthroughs in artificial intelligence over the next year. Speaking at the New York Times’ DealBook Summit, Pichai explained that while AI technology will continue evolving, it will likely experience a slower pace of advancement by 2025.
He acknowledged that while current tools like ChatGPT, Google’s Gemini, and Meta’s Llama are improving incrementally, the next big leap in AI will require more substantial breakthroughs. Pichai described the industry’s current trajectory as “a steeper hill,” meaning that progress will become increasingly challenging. Even so, these gradual improvements will make AI more reliable and accessible to a broader audience.
The Rise of Generative AI and Current Landscape
Generative AI exploded onto the scene with the launch of ChatGPT by OpenAI in late 2022, sparking a wave of innovation across industries. Since then, tech giants such as Google, Meta, and Microsoft have joined the fray, introducing their own AI models. However, Pichai notes that the initial period of rapid, game-changing development is over.
“The low-hanging fruit is gone,” Pichai remarked, explaining that significant breakthroughs will now be necessary for AI to continue growing. According to him, the kind of revolutionary advances that shook the industry in the past few years are unlikely to happen in 2025. Instead, progress will likely come through smaller, but important improvements.
AI’s Growing Impact and Financial Investment
Although AI technology is becoming increasingly powerful, it has not yet lived up to its profit-making potential for many companies. Despite significant investments, AI has yet to generate substantial revenue, even as financial reports predict spending on AI technology will surpass $1 trillion in the near future.
Pichai sees a positive side, however. As AI technology advances, it is likely to make fields like computer programming more accessible to a wider group of people. He anticipates that over the next decade, millions more individuals will have the opportunity to enter tech careers, thanks to AI. He also highlighted how AI is creating new job opportunities, with roles such as AI trainers earning more than $64,000 per year and prompt engineers making over $110,000 annually, according to ZipRecruiter.
Diverse Views on AI’s Future
Pichai’s cautious outlook is shared by other tech leaders, including Microsoft’s CEO Satya Nadella. Nadella compared AI’s evolution to the Industrial Revolution, noting that the early phases were slow before significant acceleration occurred. “It’s never going to be linear,” Nadella said during a recent appearance at the Fast Company Innovation Festival.
However, not all in the AI community agree with this perspective. OpenAI’s CEO Sam Altman publicly dismissed the idea of a slowdown, asserting that “there is no wall” to AI progress. He made this statement in response to concerns that the newest version of ChatGPT, GPT-4, only showed modest improvements over earlier models.
Ilya Sutskever, a co-founder of Safe Superintelligence and former OpenAI chief scientist, recently expressed similar sentiments, acknowledging that scaling pretraining models has begun to yield diminishing returns. “Everyone is looking for the next thing,” he said, underscoring the search for new breakthroughs.
Pichai also responded to criticism from Satya Nadella, who has questioned Google’s leadership in AI despite the company’s early start. “I would love to do a side-by-side comparison of Microsoft’s own models and our models any day, any time,” Pichai said. He pointed out that Microsoft primarily relies on OpenAI’s models rather than developing its own, signaling confidence in Google’s AI capabilities.
“You can perceive it as there’s a wall, or there are small barriers,” he said. Despite these challenges, Pichai expressed confidence that AI’s development will continue to produce meaningful results, even if the progress is more gradual than the initial surge of innovation.
