Elon Musk’s xAI is reportedly in discussions to secure funding that would push its valuation to an impressive $40 billion.
According to The Information: “Sources indicate that representatives from xAI, including Musk’s close associate Jared Birchall, have been engaging with potential investors for a major funding round. Usual investors such as Valor Equity Partners, Sequoia Capital, Andreessen Horowitz, and Vy Capital have shown interest. However, xAI might prioritize raising funds from strategic partners, potentially sidelining some traditional VC firms. So, who could be a potential strategic investor? Our bet is on Nvidia.”
Nvidia’s relationship with xAI is notably closer than its ties with others like Anthropic and OpenAI, who typically lease GPUs via cloud giants like Microsoft or AWS. Musk’s ownership of a Memphis-based data center means Nvidia can directly supply him with chips and potentially shape xAI’s procurement decisions, including networking and supercomputer equipment.
The secondary market for xAI, as well as for Anthropic, which is also in talks to secure funding at a $40B valuation, is now mostly bid-driven.
Last week brought major developments in search from both Meta and OpenAI. Meta announced plans to advance its search engine, aiming to lessen its dependency on Google and Bing. Meanwhile, OpenAI rolled out a search feature within ChatGPT.
These moves coincide with ongoing discussions about a potential acquisition of Perplexity. X, OpenAI, Notion, and Microsoft have all been mentioned as potential buyers. But could Meta be a contender?
With heightened competition in the search space, acquiring Perplexity could benefit any of these companies. However, the real question remains: would regulators allow such a deal?
OpenAI has explored various strategies to diversify its chip supply and cut costs, including the possibility of building an entire chip production operation in-house and raising funds to create a network of chip-manufacturing foundries. However, due to the significant financial and time investments required, OpenAI has decided to put the ambitious foundry plans on hold. Instead, the company is shifting focus to in-house chip design, according to sources who spoke on condition of anonymity.
OpenAI has been collaborating with Broadcom for several months to develop its first AI chip, which will prioritize inference processing. While current demand leans heavily toward training chips, analysts predict that the need for inference chips could surpass them as more AI applications roll out. The chip team, composed of about 20 experts—including Thomas Norrie and Richard Ho, known for their work on Google’s Tensor Processing Units (TPUs)—is spearheading the effort.
Through Broadcom, OpenAI has secured manufacturing capacity with Taiwan Semiconductor Manufacturing Company (TSMC) for its custom-designed chip, slated for 2026, though this timeline may shift.
While OpenAI’s moves highlight the dominance of public companies in the chip sector, how are private chip manufacturers faring? Recent developments have been mainly in areas that complement major tech firms, such as Lightmatter, Crusoe, and CoreWeave. Notably, Microsoft plans to invest nearly $10 billion by 2030 to lease servers from CoreWeave, an AI-focused startup. This figure, disclosed to investors, exceeds previous estimates and makes up over half of the $17 billion in contracts CoreWeave has secured with its clients.
Sierra Technologies Inc., an emerging leader in artificial intelligence co-founded by former Salesforce co-CEO Bret Taylor, announced the close of a $175 million funding round, pushing its valuation to $4.5 billion.
Alex Wilhelm of Cautious Optimism weighed in on the hefty valuation: “What I don’t get is the price.” According to Reuters, Sierra’s annual recurring revenue stands at approximately $20 million, equating to a valuation/revenue multiple of 225. This surpasses even the ambitious 157 multiple set by Perplexity last week.
Wilhelm remarked, “With Sierra, we’re once again witnessing a company being valued so far ahead of its current revenue that it’s effectively pre-sold years of growth and success. This means the company will need to achieve extraordinary growth just to justify its existing numbers, not to mention raise future capital at potentially higher costs.”
Productivity startup Miro announced it would reduce its workforce by 18%. Similarly, blockchain developer Consensys and publicly traded file-sharing company Dropbox each trimmed their teams by 20%, while Kraken let go of 15% of its staff. Are these moves signs that AI is replacing us?
According to PitchBook, the broader tech landscape tells a different story. Startups seem to have largely right-sized their headcount since the market correction in mid-2022, and layoffs are becoming less common.
Overall, layoffs in tech have decreased significantly. Data from Layoffs.fyi, which tracks job cuts, shows that fewer tech workers are being laid off now than at any point since Q2 2022. In October, layoffs impacted just over 3,000 employees across 33 tech companies, compared to more than 8,000 in October 2023. The number of tech layoffs has been consistently declining since its peak in Q1 2023, with Q3 seeing a 45% drop in layoff instances compared to the same period last year.
For a clearer perspective, let’s compare Miro's staffing levels with other high-liquidity private companies in the productivity sector.
Canva has announced a significant milestone, reaching 200 million monthly active users—a growth of over 115 million in the past two years. The company also reported $2.5 billion in annualized revenue, marking the first revenue disclosure of 2024 among private productivity companies. This represents nearly 50% year-over-year revenue growth, just slightly below last year's 55%, which is impressive given current market conditions. Canva’s valuation on the secondary market is now estimated at approximately $28 billion.