Artificial Intelligence Funding Latest Silicon Valley Gold Rush

Silicon Valley will be producing fewer “like” buttons following a funding shift away from social media to deep learning and artificial intelligence.

VCs’ interest in deep learning is surging. CB Insights predicts that funding of AI-enabled startups will reach $1.2 billion in 2016, compared with only $145 million in 2011, when social media startup funding peaked at $2.4 billion. This year so far, the figure stands at $6.9 million.

Why Deep Learning Took Off

Deep learning is a type of machine learning that uses layered neural networks, loosely inspired by the brain, to find patterns in huge amounts of data. The basic ideas had been around for decades, but they only started to beat older approaches once researchers had two things in quantity: data to train on and cheap computing power, especially graphics chips, to do the training.

Once those were in place, deep learning made rapid progress on problems computers had long struggled with, like recognizing images and understanding speech. That’s what turned it from an academic specialty into something investors wanted a piece of.

John Shoch, a VC at Alloy Ventures in Palo Alto, told the New York Times that new approaches in the space, such as those from Facebook head of AI Yann LeCun, have led to more investors rushing to fund the AI space: “You get a new set of tools that let you attack a new set of problems, which let you push the boundary out.”

Silicon Valley could have an opportunity to solve physical problems, instead of concentrating on social media-based apps like food delivery or dating that support “assisted living for millennials,” as Recode editor Kara Swisher alleges. Industries where new AI-enabled startups are working include radiology diagnostics, from Enlitic, and enterprise intelligence, from MetaMind, recently acquired by SalesForce.

An artificial intelligence research centre, OpenAI, was funded in December 2015 with the aim of focusing on a positive human impact, backed with $1 billion of funding. The organization’s mission is to extend “individual human wills and, in the spirit of liberty, [make AI] broadly and evenly distributed as possible.”

Still, the transition to AI in the startup space needs more depth in terms of resources.

Jana Eggers (above), CEO of AI-based personalized recommendation startup NaraLogics, says, for example, that there is still lack of diversity in the deep learning space. She told StartupFest Montreal that if only rich white males are programming AI, the technology can’t reach its full potential.

“We need people like you, and we need people not like you,” she told the audience.

From Gold Rush to Boom

The $1.2 billion forecast for 2016 looks tiny now. According to Stanford’s 2025 AI Index report, U.S. private investment in AI reached $109.1 billion in 2024 alone. Generative AI attracted $33.9 billion in private investment worldwide that year, and 78% of organizations reported using AI, up from 55% the year before.

The researchers named in this story were part of that shift. Yann LeCun, along with Yoshua Bengio and Geoffrey Hinton, received the 2018 ACM A.M. Turing Award, often called the Nobel Prize of computing, for their work on deep learning.

What Happened to OpenAI

OpenAI changed a great deal from the research centre described above. In March 2019, it restructured as a “capped-profit” company so it could raise the capital it needed, with investors’ returns limited to 100 times their investment. Then, on November 30, 2022, it released ChatGPT, the chatbot that brought AI into everyday use and set off a new wave of investment far bigger than the one this article described.

What It Means for Founders

The shift Silicon Valley was beginning in 2016, from social apps to AI, turned out to be real. But Jana Eggers’s warning has aged well too. AI systems learn from data and are designed by people, and when the people and the data are narrow, the results can be too. A team that brings different backgrounds and perspectives is better placed to spot blind spots before customers do.

For founders, the lesson is to look at AI as a tool for solving real problems, like the radiology and enterprise examples above, rather than as a label to attract funding. The money followed the hype in 2016, and it’s following it again now. The companies that last are the ones that use the technology to do something people actually need.