Qualcomm has reportedly approached Intel with the possibility of acquiring the troubled chipmaker, a move that could potentially reshape the tech ecosystem.

According to several sources, while the discussions are still at an early stage, Qualcomm CEO Cristiano Amon is personally engaged in exploring the acquisition. Amon and his team are evaluating Intel’s portfolio, including its PC design unit, which has been of particular interest.

Intel’s Struggles

Intel which was once a titan of the chip industry has seen its market value decline by almost 60% this year, with its shares struggling amid mounting competition and internal challenges.

Intel has also struggled to keep up in the AI era despite being a dominant force in the 1990s and 2000s and has notably fallen behind Taiwanese rival TSMC in manufacturing, as well as competitors like Nvidia and AMD, which have capitalized on the generative AI boom.

Simultaneously, Intel’s decision to focus on AI processors and turn its business around by building a foundry (a chip contract manufacturing business) hasn’t yielded significant results yet.

The company has also paused major factory construction projects in Europe and scaled back its real estate holdings. A recent deal with Amazon to manufacture custom networking chips may offer a lifeline, but Intel’s woes remain largely unresolved.

Intel’s Missed Opportunities

One of Intel’s biggest missteps has been its rejection of a strategic partnership with OpenAI.

Intel had the chance to purchase a 15% stake in OpenAI for $1 billion between 2017 and 2018, which would have given Intel a foothold in the AI space.

The chipmaker also discussed the possibility of a second 15% stake if it manufactured hardware for OpenAI at cost. However, then-CEO Bob Swan dismissed the opportunity, believing that generative AI models wouldn’t bring quick returns.

Hence, Intel’s decision not to invest in OpenAI, which later developed ChatGPT and reached a valuation of around $80 billion, now seems like a colossal oversight.

This miscalculation has resulted in Intel’s staying competitive, the company has been slow to market with AI chips, and its stock has suffered, reaching its lowest point since 1974.

Intel’s value dipped below $100 billion for the first time in three decades while Nvidia, riding the AI wave, has surged to a $2.6 trillion valuation.

The Qualcomm Approach

Qualcomm’s market value of $188 billion, seemingly wants to take advantage of Intel’s vulnerable position.

Even though no formal offer has been made, Qualcomm’s interest is reportedly strong, yet the acquisition is likely to face several hurdles including potential antitrust scrutiny from regulators in the U.S., China, and Europe.

However, if a deal does proceed, Qualcomm may have to divest parts of Intel to secure approvals. Financially, Qualcomm has $13 billion in cash, but acquiring Intel, valued at $122 billion (including its debt), would require significant resources.

Moreover, Qualcomm lacks the experience Intel has in chip manufacturing, which could be a challenge as Intel has invested billions in its fabrication process over decades.

Implications for the Tech Industry

If Qualcomm successfully acquires Intel, it would mark one of the largest deals in tech history since Broadcom’s attempted $142 billion acquisition of Qualcomm in 2018, which was blocked by the Trump administration.

The acquisition would create ripples across the chipmaking sector, potentially reshaping the competitive dynamics between Qualcomm, Nvidia, AMD, and TSMC.

Intel’s Big Losses

It is among a series of strategic misfortunes that have seen the company, which was at the cutting edge of computer chips in the 1990s and 2000s, stumble in the era of AI.

Intel’s second-quarter earnings resulted in a stock price collapse, marking its worst trading day since 1974, with more than a quarter of its value wiped out. For the first time in 30 years, the once-unstoppable tech giant is now worth less than $100 billion.

Intel, once synonymous with innovation—its “Intel Inside” slogan representing the pinnacle of chip quality—continues to stutter in the AI revolution.

Meanwhile, rival Nvidia, with a market valuation of $2.6 trillion, has transformed its business from video game graphics chips to AI-focused hardware that powers generative AI models like OpenAI’s GPT-4 and Meta’s Llama models.

AMD, another competitor, is now valued at $218 billion, widening the gap further between Intel and the industry’s frontrunners.

Damage Control

Intel’s sluggish response to the AI boom has left it scrambling to catch up.

CEO Pat Gelsinger has pointed to the company’s third-generation Gaudi AI chip, expected to launch in the third quarter, as the company’s answer to Nvidia and AMD.

Gelsinger noted that Intel already has over 20 customers for its second and third-generation Gaudi chips. Likewise, Intel’s next-generation AI chip, Falcon Shores, is slated for a 2025 release.

“We are nearing the completion of a historic pace of design and process technology innovation,” said a company spokesperson, expressing optimism about Intel’s AI strategy.

Gaming Chips Dominate AI

Intel’s missed opportunity with OpenAI in 2017 was a turning point, but its troubles in AI go much deeper.

In 2019, Microsoft filled the void Intel left by investing in OpenAI, positioning itself as a leader in the AI-driven tech landscape. The 2022 launch of OpenAI’s ChatGPT sparked a massive AI frenzy, pushing Intel further behind the competition.

Though Intel’s decision to pass on OpenAI may appear glaring in retrospect, the company’s struggle for AI dominance had been brewing for over a decade, according to experts.

“Intel failed in AI because they didn’t present a cohesive product strategy to their customers,” remarked Dylan Patel, founder of SemiAnalysis, a semiconductor research firm.

For over 20 years, Intel firmly believed that its CPU (central processing unit) architecture—the kind found in most desktops and laptops—could handle the data processing required for AI tasks.

According to four former Intel executives, the company underestimated the potential of GPUs (graphics processing units), which were being used by Nvidia and AMD for video gaming.

Intel engineers viewed GPUs as inefficient and “ugly” compared to CPUs, according to sources. However, by the mid-2000s, researchers had discovered that GPUs were much better suited for the intensive parallel processing necessary to build and train large AI models.

Unlike CPUs, which process a few operations at a time, GPUs can perform thousands of calculations simultaneously—a game-changer in AI.

Nvidia, seeing the potential, spent years refining its GPU architecture to cater to AI-specific tasks, building out the software to harness its capabilities.

“When AI hit… Intel just didn’t have the right processor at the right time,” said Lou Miscioscia, an analyst at Japanese investment bank Daiwa.

Nervana and Habana, Intel’s AI Chip Struggles

Since 2010, Intel has made multiple attempts to gain a foothold in the AI chip market, including acquiring startups and launching several internal efforts. Unfortunately, none of these initiatives have made a meaningful impact in the face of competition from Nvidia and AMD.

According to three individuals familiar with Intel’s operations, the company’s investments in AI hardware have yet to deliver significant returns. Despite Intel’s efforts, its entire data center business, which includes AI chips and other designs, is only expected to generate $13.89 billion in sales this year.

In contrast, Nvidia’s data center revenue is projected to reach an astronomical $105.9 billion, stressing Intel’s inability to compete in this highly lucrative sector.

In 2016, under the leadership of CEO Brian Krzanich, Intel sought to buy its way into the AI revolution by acquiring Nervana Systems for $408 million.

Nervana’s technology, much like Google’s tensor processing unit (TPU), was designed specifically for AI workloads, stripping away features meant for gaming and focusing on optimizing AI computations.

Nervana had some success, even securing Meta Platforms as a customer, but it wasn’t enough. By 2020, Intel abandoned the Nervana project in favor of another acquisition.

In 2019, Intel acquired Habana Labs for $2 billion and subsequently shut down Nervana’s operations. Habana Labs was seen as a better fit for Intel’s long-term AI strategy, but as of now, it has not yet propelled Intel into the upper echelons of the AI hardware market.

How Intel Lost the Sony PlayStation Business

In 2022, Intel suffered another major setback when it lost out on a bid to design and fabricate the chip for Sony’s highly anticipated PlayStation 6 console.

The loss was a severe blow to Intel’s contract manufacturing business, which was a key part of CEO Pat Gelsinger’s vision to turn the company around.

If Intel had secured the PlayStation 6 chip it would have not only strengthened Intel’s design segment but also served as a major win for its new foundry business.

The bidding process for the PlayStation 6 contract was fiercely competitive, with Intel and AMD emerging as the two final contenders.

The contract promised billions of dollars in revenue, with the fabrication of thousands of silicon wafers per month. Winning the contract would have meant steady business for Intel, as Sony typically sells more than 100 million units over the lifecycle of its gaming consoles.

Although the console chip business has lower profit margins compared to products like AI chips, it still represents a steady stream of revenue that can be profitable with already-developed technology.

The PlayStation deal could have been a much-needed boost for Intel’s foundry business, which has struggled to secure major clients. However, a pricing dispute ultimately caused the deal to fall through.

Two sources indicated that Intel and Sony were unable to agree on how much profit Intel would earn from each chip sold to the Japanese electronics giant. AMD, which had been competing fiercely with Intel, ultimately secured the contract.

The loss was particularly disappointing given the extensive negotiations that took place between Sony and Intel throughout 2022.

The discussions involved meetings between both companies’ CEOs, dozens of engineers, and high-level executives. In the end, however, AMD emerged victorious, securing yet another high-profile win in a market where Intel once dominated.

Backwards Compatibility and Sony’s PlayStation Deal

The current generation of Sony’s PlayStation consoles is powered by custom chips designed and manufactured by AMD, including the PlayStation 5, which launched in 2020 and has sold over 20.8 million units by fiscal 2023.

Recently, Sony announced the PlayStation 5 Pro, but the next-generation console remains unrevealed. For Sony, continuity is essential—custom chips for consoles often ensure backward compatibility, allowing users to play older games on new hardware.

This was a significant point of discussion between Intel and Sony engineers during the PlayStation 6 contract talks.

Switching from AMD, which designed the PlayStation 5 chip, to Intel would have risked disrupting that compatibility, presenting a technical challenge that required significant engineering resources.

Backwards compatibility has long been a staple for Sony’s gaming systems, and maintaining this feature in new systems is not just a technical necessity but a business imperative.

Despite these discussions, Intel ultimately missed out on securing the PlayStation 6 contract while AMD continued its relationship with Sony.

Intel’s failure to secure the deal compounded its struggles in the chip-making space, especially as it tries to regain its footing amid growing competition in the AI sector from Nvidia and AMD.

Intel’s Struggles in AI and Manufacturing

Having already missed the first wave of the AI boom, Intel has been facing severe financial challenges.

In August, the company reported its worst second quarter in decades and announced plans to lay off 15% of its workforce to save $10 billion.

Additionally, Intel scaled back its ambitious foundry expansion, a key element of its strategy to compete in chip manufacturing.

At the same time, internal instability became evident with the sudden departure of high-profile board member Lip-Bu Tan. His exit, attributed to differences in Intel’s strategic direction, added pressure on CEO Pat Gelsinger.

As part of its ongoing review, Intel’s leadership has been discussing potential divestitures of businesses that are no longer deemed viable, including the future of its programmable chip unit, Altera, and its manufacturing expansion in Germany.

The split of Intel’s design and manufacturing units under Gelsinger’s leadership has not yet borne fruit, as the company struggles to find a marquee customer for its advanced 18A manufacturing process.

The 18A process, crucial to Intel’s foundry strategy, has so far failed to attract major customers, which is a major setback for the company’s ambitions to rival Taiwan Semiconductor Manufacturing Company (TSMC).

The Missed PlayStation 6 Opportunity

Intel’s failure to win the PlayStation 6 chip design and manufacturing contract was a massive missed opportunity, particularly for its fledgling foundry business. Had Intel secured the deal, it could have occupied its foundry unit for over five years, injecting much-needed revenue into the company.

According to Intel’s internal projections, the PlayStation 6 contract could have generated approximately $30 billion over the contract’s lifetime, based on the success of Sony’s past consoles like the PlayStation 2, which sold roughly 150 million units since its launch in 2000.

Securing Sony’s console business would have not only provided a stable and long-term source of revenue but also elevated Intel’s profile as a contract manufacturer, potentially attracting more marquee clients.

The loss of the PlayStation contract adds to Intel’s challenges as it continues to struggle in its pursuit of high-profile customers for its advanced 18A process, a critical piece of Gelsinger’s turnaround plan for the company.