Intel Missed Key Opportunity with OpenAI, Struggles to Keep Pace in AI Era

326

Several years ago, Intel, once the leader in computer chips, missed a crucial opportunity that could have dramatically altered its trajectory in the AI era. Around 2017-2018, the tech giant had the chance to invest in OpenAI, a nascent organization at the forefront of a field then known as generative artificial intelligence. Four individuals with direct knowledge of these discussions shared this information with Reuters.

During these discussions, which took place over several months, Intel considered several investment options, including acquiring a 15% stake in OpenAI for $1 billion in cash. Additionally, there were talks about Intel potentially securing an extra 15% stake if it agreed to manufacture hardware for OpenAI at cost price. Despite these negotiations, Intel ultimately chose not to proceed with the investment.

The decision was influenced by the skepticism of Intel’s then-CEO, Bob Swan, who doubted that generative AI models would soon become commercially viable or justify the investment. Sources who requested anonymity revealed that Intel’s data center division was also hesitant about manufacturing products at cost, which further contributed to the collapse of the deal.

A spokesperson from Intel declined to comment on the discussions regarding the potential investment in OpenAI. Bob Swan did not respond to requests for comment, and OpenAI also chose not to provide a statement.

Intel’s reluctance to invest in OpenAI, which later launched the groundbreaking ChatGPT in 2022 and is now valued at approximately $80 billion, is a significant chapter in a series of strategic missteps. These decisions reflect a broader trend of the company struggling to maintain its leadership in an era dominated by artificial intelligence, as confirmed by interviews with nine individuals, including former Intel executives and industry experts.

The consequences of Intel’s decision were starkly highlighted last week when the company’s second-quarter earnings led to a dramatic 25% drop in its stock price, marking its worst trading day since 1974. For the first time in three decades, Intel’s market capitalization fell below $100 billion. Once synonymous with cutting-edge technology and quality—symbolized by its “Intel Inside” slogan—the company is now grappling to bring a competitive AI chip product to market.

Intel now faces significant competition from Nvidia, which has emerged as a dominant force in the AI chip sector. Nvidia’s transition from video game graphics to specialized AI chips for training and operating large generative AI systems, such as OpenAI’s GPT-4 and Meta Platforms’ Llama models, has placed it in a leading position. Intel also lags behind AMD, which is valued at $218 billion.

In response to queries about its AI progress, Intel’s spokesperson pointed to recent comments by CEO Pat Gelsinger. Gelsinger highlighted the upcoming launch of Intel’s third-generation Gaudi AI chip, expected in the third quarter of this year, which he claims will outperform competitors. The company also plans to introduce its next-generation Falcon Shores AI chip by late 2025.

“We are nearing the completion of a historic pace of design and process technology innovation,” the spokesperson stated. “We are encouraged by the product pipeline we’re building to capture a greater share of the AI market going forward.”

While Intel’s missed opportunity with OpenAI represents a significant strategic error, it also reflects a broader trend of the company struggling to adapt to the rapid advancements in AI technology. Microsoft stepped in to invest in OpenAI in 2019, significantly boosting its position in the AI landscape following the release of ChatGPT in 2022 and igniting a flurry of activity among leading tech companies to deploy AI technologies.

According to former Intel executives and industry analysts, Intel’s failure in the AI domain can be attributed to a lack of cohesive product strategy. For over two decades, Intel focused on central processing units (CPUs) for AI tasks, underestimating the potential of graphics processing units (GPUs) used by rivals like Nvidia and AMD. GPUs, initially designed for video game graphics, proved to be far more efficient for handling the data-intensive tasks required for AI model training and development.

Nvidia has spent years optimizing its GPU architecture for AI applications and developing the necessary software to harness its capabilities. Intel, on the other hand, struggled to adapt its technology to meet the evolving demands of the AI industry.

Since 2010, Intel has made several attempts to develop a competitive AI chip, including acquiring startups and launching internal projects. Despite these efforts, Intel has struggled to make a significant impact in the AI chip market. In 2016, Intel acquired Nervana Systems for $408 million, attracted by its technology similar to Google’s tensor processing units (TPUs). However, the project was eventually abandoned in favor of acquiring Habana Labs for $2 billion in 2019.

Intel’s data center business, including its AI chips, is projected to generate $13.89 billion in sales this year, while Nvidia’s data center revenue is expected to reach $105.9 billion. This stark contrast underscores the challenges Intel faces in regaining its footing in the rapidly growing and lucrative AI market.

As Intel continues to navigate its path forward, the lessons from its past decisions and the competitive landscape will play a crucial role in determining its future success in the AI era.