TheChek News Flow

AMD’s investment is part of how the chip giant plans to make a dent in the artificial intelligence chip market dominated by Nvidia

Chip giant Advanced Micro Devices and hedge fund LuminArx Capital Management raised $333 million for a growth financing round for cloud startup Vultr, a sign of the hot demand for artificial intelligence infrastructure.

West Palm Beach, Fla.-based Vultr said it is now valued at $3.5 billion and plans to use the financing to acquire more graphics processing units, or GPUs, the chips that power AI models. The funding is Vultr’s first injection of outside capital.

The company got its start providing cloud-computing for businesses’ information-technology systems and it also offers AI computing. Vultr said its AI cloud service, in which it leases GPU access to customers, will soon become the biggest part of its business.

For AMD, the investment is a way for “customers to experience AMD GPUs through their platform,” said Mathew Hein, AMD’s chief strategy officer of corporate development. “We look at them as somebody that’s going to be bringing up AMD capacity this generation and future generations.”

AMD aims to become Vultr’s “preferred” AI hardware provider, though it won’t force the choice, Hein said. Earlier this year, AMD unveiled its latest generation of AI chips, called the MI325X, with a next generation of MI350 chips coming next year.

Earlier this month, Vultr announced plans to build its first “supercompute” cluster with thousands of AMD GPUs at its Chicago-area data center.

The funding round follows similar moves by AI chip market leader Nvidia, which alongside other investors, provided more than $400 million to AI cloud provider CoreWeave in 2023. CoreWeave last year also secured $2.3 billion in debt financing by using its Nvidia GPUs as collateral.

Nvidia has spread out access to its GPUs to a large number of cloud providers, as has AMD.

“AMD is trying to figure out how to become more competitive with Nvidia,” said Dave McCarthy, a research vice president in cloud and edge services at research firm International Data Corp. “For AMD to be able to get good billing with an up-and-coming cloud provider like Vultr will help them get more visibility in the market.”

AMD has also invested in cloud providers such as TensorWave, which also offers an AI cloud service. In August, AMD bought the data-center equipment designer ZT Systems for nearly $5 billion.

The market for AI semiconductors will reach an expected $193.3 billion by the end of 2027 from an estimated $117.5 billion this year, according to IDC. Nvidia commands about 95% of the market for AI chips, according to IDC.

Vultr said its cloud platform serves hundreds of thousands of businesses, including Activision Blizzard, the Microsoft-owned videogame company, and Indian telecommunications giant Bharti Airtel.

Vultr’s customers also use its decade-old cloud platform for their core IT systems, said Chief Executive J.J. Kardwell.

There is no shortage of entrants into the AI chip market aiming to challenge Nvidia’s lead. In addition to AMD, startups like SambaNova Systems, for instance, and cloud giants Amazon Web Services and Google seek to offer alternatives to Nvidia’s hardware.

Microsoft, Meta Platforms and Oracle have said they use AMD’s GPUs. A spokesperson for Amazon’s cloud unit said the company works closely with AMD and is “actively looking at offering AMD’s AI chips.”

Like most cloud platform providers, Vultr isn’t sticking with one GPU supplier. It offers Nvidia and AMD GPUs to customers, and plans to keep doing so, Kardwell said. “There are different parts of the market that value each of them,” he added.

The part of the market using AI models rather than training them, deploying a process called AI inference, requires faster speeds and more bandwidth. That plays to Vultr’s strengths and is an area where AMD’s GPUs are competitively priced, Kardwell said.

Like CoreWeave, a New Jersey-based company that got its start in cryptocurrency mining, Vultr is part of a wave of cloud companies getting chips from and being funded by chip makers. Both companies also quickly pivoted to acquiring GPUs to meet the demand for generative AI, said Lee Sustar, an analyst at Forrester.

Sam Altman-led nuclear start-up signs major AI power supply deal

Contract is the latest between energy developers and technology industry amid soaring demand

Oklo, a nuclear energy start-up chaired by Open AI’s Sam Altman, has struck a major corporate power supply deal as the industry rushes to meet the surging needs of artificial intelligence.

The 20-year agreement with Switch Inc, a large privately held data centre operator, is to build reactors with a total capacity of up to 12 gigawatts — enough in total to power all 7.6mn households in New York state.

Oklo claimed the agreement was among the largest clean power deals in history, even though it is non-binding and the company’s technology is years from production.

Jacob DeWitte, Oklo’s co-founder and chief executive, told the Financial Times that if binding terms are reached, the deal could be worth multiple billions of dollars.

The contract is the latest announced between nuclear developers and the technology industry, as the artificial intelligence boom creates a dire need for high-wattage, low-carbon energy supplies.

Oklo is developing small modular reactors — new types of advanced nuclear plants with a capacity of 300 megawatts or less, which is about a third of standard facilities.

Big Tech is increasingly betting that small reactors can deliver enough energy to run AI systems, even though no western company has yet successfully deployed the technology.

DeWitt said nuclear was the only power source that could sustainably meet the massive energy demands of the AI revolution. He downplayed critics’ concerns about the risks inherent in nuclear energy, which is highly regulated and has been subject to delays and cost overruns.

“We are entering a new world due to the size of the energy demand,” DeWitte said in an interview. “You can’t do it with renewables, as you would need a lot of gas back-up capacity and a lot of folks want a clean solution.”

Oklo is aiming to deploy its first 15MW reactor by late 2027 at the Idaho National Laboratory. Rival US-based nuclear developers X-energy and Kairos Power recently signed power deals with Amazon and Google to provide low-carbon electricity to power their data centres.

Microsoft also struck a corporate power agreement in September with Constellation Energy that will reactivate the Three Mile Island nuclear plant in Pennsylvania.

Adam Stein, director of nuclear energy innovation at The Breakthrough Institute, a Washington think-tank, said Big Tech’s contracts with nuclear developers would give investors confidence to support a nascent industry.

“They are intentionally taking on some of that technology risk, the first-mover risk in their power-purchase agreements,” he said. “This is how new technology gets to market.”

But some analysts are sceptical that the new crop of nuclear start-ups will ever deliver.

“These agreements do not appear to be the kinds of serious, substantial, and sustained financial commitments — on the order of many billions of dollars over decades — that would be necessary to fully realise these speculative nuclear projects,” said Edwin Lyman, nuclear power safety director for the Union of Concerned Scientists.

Although nuclear power generation does not emit carbon dioxide, some critics deny it is a clean power source, pointing to the radioactive waste it creates, which must be stored indefinitely.

Switch operates several large-scale data centres in Nevada, Texas and Atlanta using renewable power. To fulfil the maximum terms of the contract, Oklo would have to build hundreds of its small reactors across the US to serve Switch’s facilities.

Oklo, which counts technology entrepreneurs Altman and Peter Thiel among its early investors, listed in New York in March and has a $2.2bn market capitalisation.

Chris Wright, Donald Trump’s nominee to become US energy secretary, is a director of the company. DeWitte said Wright would step down from the board if he is confirmed by the US Senate.

Carmakers face problems of Chinese competition, a shrinking domestic market and the threat of tariffs

Merger talks between Honda and Nissan and the potential creation of the world’s third-biggest carmaker represent a critical admission for the whole of corporate Japan: the best time to consolidate was yesterday, the second best time is today.

That was the view of one senior government official on Wednesday, in a reflection of deepening concerns over the survival of Japan’s fragmented automotive industry and the collapse of Nissan’s market value.

The talks to close ranks and combine are taking place in a hostile environment. Chinese competitors are relentless, the tariff and global trade regime under Donald Trump is unpredictable, and Japan’s economy is swapping years of ultra-loose monetary policy for rising interest rates.

The response of the nation’s second and third-biggest carmakers will reverberate throughout the economy, say analysts. A decision to consolidate could force hundreds of other Japanese companies in other sectors to look around them and decide that dealmaking may be the only route to survival.

Japan’s automotive industry is coming under assault from China’s affordable, slick electric vehicles even as it faces the rising threat of tariffs on exports to the US where brutal discounting has eroded profitability for all but the biggest producers.

Linked to Nissan and Honda are a vast web of suppliers and industrial companies, many making the same products — from ball bearings and lifts to semiconductors — while facing ever stiffer competition from China.

As well as growing global competition, Japan Inc is being pushed towards mergers by investor-friendly corporate governance reforms, rising shareholder activism, a shrinking domestic market and tightening labour availability, say the nation’s top executives.

Takeshi Niinami, chair of the Japanese Association of Corporate Executives, said that the mindset towards consolidation in Japan was shifting as the country moved to a new era of inflation following three decades of stagflation.

“Consolidation is burgeoning in this country and I think we should just see more,” said Niinami, who is also chief executive of Suntory Holdings, the beverage company. “Now is the right time.”

Nissan serves as an example of how Japanese brands have fallen down the global pecking order. In 2013, the company was the sixth largest automaker in the world, selling 4.9mn cars. This year, it only expects to sell 3mn and has suffered in the US market from a lack of hybrid offerings, which have surged in popularity and sheltered Toyota’s financials.

Honda has also shrunk over that time, going from 4.3mn automobiles a decade ago to an expected 3.8mn this year.

By contrast, Chinese rivals such as BYD have grown into some of the world’s largest selling brands on the back of early investment in electric vehicle technology and big economies of scale.

The industry still faces more vast investment outlays for battery technology and software — areas where China’s carmakers, which have benefited from years of Beijing support towards technical research and securing global supply chains for key resources, have an edge over the traditional engine-based expertise of their Japanese rivals.

Jeff Hutchins, head of Japan equities at Jefferies, said Japan was in the early stages of what was likely to be a huge, multiyear increase in corporate activity and improvements in capital efficiency. This would be led by M&A activity and driven by the increasing pressures on traditional automakers.

“The US and Germany have shown the playbook for autos consolidation, now it is Japan’s turn to follow,” he said.

Nissan and Honda have been exploring a partnership since March last year, taking a further step of announcing joint collaboration on EVs and software over the summer. Honda also agreed last year to partner with Sony to pool resources in vehicle engineering and software to create cars together.

A person close to Nissan said the company had been in an alliance with Renault for about 20 years without the desired effects, and antitrust regulations limited the depth of collaboration and information, creating the need to explore a merger.

Masatoshi Kikuchi, pan-Asian chief equity strategist at Mizuho, said that Japanese auto companies faced a trio of problems as they lost market share to Chinese rivals in that country and south-east Asia, while facing a shrinking domestic market and additional tariffs from Trump.

“Japanese car companies, especially Nissan and Honda, decided to discuss merging as they need to tackle several headwinds at once,” he said.

Kikuchi cast doubt on whether Nissan and Honda merging would cause a ripple effect throughout Japan, citing management teams’ strong desire to remain independent — until they come under attack from activist investors.

Nissan itself has welcomed two such investors, Effissimo Capital Management and Oasis Management, to its register.

Please use the sharing tools found via the share button at the top or side of articles. Copying articles to share with others is a breach of FT.com T&Cs and Copyright Policy. Email licensing@ft.com to buy additional rights. Subscribers may share up to 10 or 20 articles per month using the gift article service. More information can be found at https://www.ft.com/tour.

https://www.ft.com/content/696b4d53-1415-47ae-a832-b2a6ceb4cc77

Executives at Nissan believe the Chinese market is rapidly being ceded to local players, despite it formerly being a stronghold. Production cuts in the country will play only a small role in helping Nissan to achieve 9,000 job cuts, as outlined in a restructuring plan in November, because of the joint venture structures with local partners that Beijing mandates.

For the Japanese carmaker, the biggest issue is the US market, in which only cash-rich companies such as Toyota and Hyundai can withstand the price-discounting war, and Nissan’s footprint has fallen behind Stellantis, General Motors and Ford.

Tariffs of up to 25 per cent from Trump on vehicle exports from Mexico threaten to hit Honda and Nissan. That makes bulking up domestic production in the US imperative, analysts say.

At the same time, Japanese automakers weak in the EV sector could struggle when some US states led by California introduce stricter emissions legislation from 2026, where hybrids’ capabilities are not enough.

Kota Yuzawa, analyst at Goldman Sachs, said two Japanese auto industry groups, led by Toyota with about 15mn in sales and another spearheaded by Nissan and Honda with 10mn sales, would have sufficient economies of scale, assuming US decoupling from China continues.

“Even so, Japanese automakers will need to keep a huge share of global hybrid sales to maintain Japan’s engine plants,” he warned, in an interview before the merger news emerged.

Nissan and Honda have significant overlap, with production concentrated in the US and Japan. That creates big potential for the two companies to reduce fixed costs.

But Japan’s demographic situation — with a shrinking workforce and consequently tight labour market — has created a new environment for domestic consolidation that would not have existed in the past, said Nicholas Smith, Japan strategist at CLSA Securities.

Japanese labour laws have traditionally made it hard to sack employees, meaning that one of the main attractions of mergers in other countries — the cost-cutting opportunities — has not been a driving force in Japan.

“You can do consolidation now because there isn’t the labour excess any more,” said Smith.

Backlog in electricity network connections has been holding back renewable energy projects

Britain’s electricity transmission network owners have set out plans to invest up to £77.4bn between 2026 and 2031 in a boost to the UK government’s clean power targets.  

National Grid, which owns the transmission network in England and Wales, has submitted plans to the regulator to invest up to £35bn over the period, it said on Wednesday.

ScottishPower Energy Networks, which covers transmission for central and southern Scotland, said it planned to invest £10.5bn, while SSEN Transmission, which covers northern Scotland, said earlier this month it planned to spend up to £31.7bn. 

It marks a big step up in investment, which will help the UK meet its target of decarbonising the electricity system by 2030.

Electricity networks need reinforcing, expanding and upgrading in order to be able to move power from growing numbers of wind and solar farms to consumers. 

A huge backlog in connection requests has built up over the past few years, raising concerns that limited network capacity is holding back renewable energy development and wider economic growth. 

In a report last month, the government’s state-owned National Energy System Operator said that, in order to reach the 2030 target, twice as much transmission network would need to be built over the next five years as had been developed over the past decade. 

John Pettigrew, chief executive of National Grid, said the plans were “unprecedented” and represented the “most significant step forward in the electricity network that we’ve seen in a generation”.

However, the plans risk raising questions over the effect on Britain’s consumer energy bills, which include a charge to fund the networks.

National Grid said its planned transmission investment would account for about £40 per year of annual household bills, up from about £20 per year at present. 

Scottish Power said its plans would account for about £12.07 a year of energy bills, up from £5.60 at present. Higher network costs should be offset by a decrease in payments to compensate generators asked to switch off at certain times due to limited network capacity.

Meanwhile, residents in several areas are protesting against new pylons and electricity cables in their neighbourhoods, raising questions over whether some plans will be blocked.

Nicola Connelly, chief executive of ScottishPower Energy Networks, said the investment would “help to stabilise and lower consumer energy bills in the longer term”.

It comes as the government is poised to launch a review of Ofgem, the energy regulator, later this week. 

Government figures said the energy department wants to overhaul the regulator to make sure it can better hold companies to account for wrongdoing and force them to raise their standards. 

The review is also expected to propose automatic customer compensation for corporate failures in the energy market. 

On Tuesday, the government brought forward to next year a planned decision on whether it should support the use of hydrogen in home heating. 

The fuel has been considered as a lower carbon alternative to the gas-fired boilers that at present heat the vast majority of homes, alongside electric heat pumps. 

The former Conservative government said it would make the call in 2026 after trials had taken place. 

But the Labour government said it wanted to “provide strategic clarity on decarbonising home heating as soon as possible” and would consult on the matter next year. 

“The government already supports existing low carbon heat options like heat pumps and heat networks, which will be the primary means of decarbonisation for the foreseeable future,” it added.

Databricks just closed one of the largest funding rounds ever, raising a staggering $10 billion in fresh capital. Naturally, technology investors were quick to ask what this means for the company’s highly anticipated IPO. During an event in San Francisco on Tuesday night, Databricks CEO Ali Ghodsi explained why he’s waiting until at least 2025 to go public.

“This year was an election year. We wanted to get some stability – people are worried about interest rates, inflation… So we said look, it’s dumb to IPO this year, so we’re definitely going to wait,” said Ghodsi during an interview with Dan Primack during the Axios AI Summit. “The earliest theoretical possibility of an IPO would be next year, and then there’s lock-up periods, so it would just be too long of a period for employees to get liquidity.”

Databricks is using this “Series J” to let early employees cash out and continue growing. While 2024 was uncertain in many ways, the IPOs of ServiceTitan, Reddit, and other companies have largely been successful.

But why risk it when you can raise as much money as Databricks?

Ghodsi said this latest round could have been nearly double the amount it just closed. We knew investors were clamoring to get in, but the craze caused Databricks to raise its share price. The data analytics company started out trying to raise $3 billion to $4 billion in this round, according to Ghodsi, but he says that press reports about their fundraising efforts drove interest through roof.

“I saw this Excel sheet where they keep a tally of all the people that want to invest. It was $19 billion of interest, and I almost fell off the chair,” said Ghodsi. “And we hadn’t even talked to everybody. I was like, ‘Oh my God, that’s a huge amount of numbers.’ So then we actually moved the price up.”

Even after the impressive fundraise, Ghodsi isn’t ruling out a Databricks IPO in 2025. However, he said it could be 2026 as well. He said it’s far less important to go public than it was 10 to 15 years ago, as this record-breaking round indicates, but it’s still something the company wants to do. That said, Ghodsi isn’t trying to squeeze in an IPO before the “AI bubble,” as he called it, bursts.

“I mean, it’s peak AI bubble. It doesn’t take a genius to know that a company with five people which has no product, no innovation, no IP – just recent grads – [is not] worth hundreds of millions, sometimes billions,” said Ghodsi. “You get billion-dollar valuations on these startups that have nothing – that’s a bubble.”

The Databricks CEO didn’t clarify what startups he was talking about, but we’ve certainly seen a lot of AI unicorns this year. None of this seems to worry Ghodsi, however, who says his company and its valuation can stand the test of time. He thinks his company has already won out its first major battle with another data analytics startup, Snowflake.

“We had a program called ‘SnowMelt,’” said Ghodsi, confirming reports of an initiative within Databricks to steal business from Snowflake. “We were going after Snowflake and we demonized them, but that’s behind us.”

That effort to demonize Snowflake came at a hefty price, reportedly causing Databricks to pay $2 billion to acquire a tiny startup called Tabular. Snowflake also reportedly wanted to buy Tabular, even though the company was only doing $1 million in annual recurring revenue at the time.

Now, Databricks is chasing bigger competitors with products that rival enterprise giants like Salesforce and Microsoft. Ghodsi says data and AI will continue to play a slightly more important role in people’s lives every year, and he thinks his company is well-positioned to fill that niche.

DAX:

MDAX:

SDAX: