M&A has come roaring back in 2026. Global dealmaking reached roughly $2.8 trillion in the first half of the year, up 49% from the same period in 2025, with a record 47 transactions worth more than $10 billion. Yet the more interesting story is not simply that companies are spending more. It is what they are spending that money on. AI, electricity, infrastructure, consolidation and access to strategic assets are increasingly sitting behind the world's biggest transactions.
At the very top of the table sits one of the strangest deals of the year: SpaceX's acquisition of Elon Musk's artificial intelligence company xAI. Announced in February, the transaction valued SpaceX at around $1 trillion and xAI at approximately $250 billion, giving the combination a headline value of roughly $1.25 trillion and making it the largest M&A transaction in history. Sullivan & Cromwell and Gibson Dunn advised on the transaction, with Sullivan & Cromwell representing xAI and Gibson Dunn acting for SpaceX.
The deal makes sense strategically, at least from Musk's perspective. SpaceX has rockets, satellites, energy and increasingly ambitious computing infrastructure, while xAI brings Grok and a rapidly expanding AI operation. Combining the two creates a vertically integrated business capable of connecting AI models, data centres, satellites and potentially future space-based infrastructure. It is difficult to separate the transaction from the enormous enthusiasm surrounding AI, however. xAI was valued at $250 billion despite being a comparatively young company whose costs are driven heavily by chips, data centres and energy. SpaceX is therefore not simply buying an existing profitable business; it is making a huge bet on what AI, space and computing infrastructure could become.
The next major deal shows that this AI infrastructure theme is not limited to technology companies. NextEra Energy agreed in May to acquire Dominion Energy in a $67 billion all-stock transaction, creating an energy company with an enterprise value of roughly $420 billion. Kirkland & Ellis advised NextEra, while McGuireWoods advised Dominion. The transaction will create one of the world's largest electricity infrastructure businesses, serving around 10 million utility customer accounts.
At first glance, an enormous merger between two regulated utilities seems a long way removed from ChatGPT. It is not. AI data centres consume extraordinary amounts of electricity, and the rapid build-out of computing capacity is beginning to turn power generation and grid infrastructure into strategic assets. The proposed NextEra-Dominion combination is therefore part of a much wider race for the physical infrastructure needed to support the digital economy. The interesting point is that companies do not necessarily need to acquire an AI company to make an AI-related investment. Sometimes the better bet is to own the electricity powering it.
The deal was not entirely unexpected given the pressures facing the utilities industry, but its scale certainly was. An important detail is that Dominion was approached by another potential bidder during negotiations, suggesting that the strategic value of large-scale electricity assets is becoming increasingly obvious. The transaction also demonstrates why today's M&A cannot always be understood simply by looking at the industry of the companies involved. Energy, technology and infrastructure are becoming increasingly intertwined.
Then there is the $110 billion acquisition of Warner Bros. Discovery by Paramount Skydance, announced in February. Paramount agreed to acquire WBD for $31 a share in cash, with the deal valuing WBD at approximately $81 billion in equity value and $110 billion including debt. The legal teams are about as heavyweight as you would expect: Latham & Watkins and Cravath, Swaine & Moore advised Paramount, while Wachtell, Lipton, Rosen & Katz and Debevoise & Plimpton advised Warner Bros. Discovery.
Unlike the SpaceX-xAI transaction, this one is much easier to explain commercially. Traditional television is under pressure, streaming businesses have struggled to produce attractive economics and entertainment companies are discovering that scale matters. Combining Warner Bros.' enormous library of intellectual property, including HBO, Harry Potter, DC and Game of Thrones, with Paramount's studios, streaming operations and distribution capabilities is an attempt to create a business large enough to compete with Netflix and the other technology-driven entertainment platforms. Paramount estimates more than $6 billion of synergies from the combination.
It is also one of the clearest examples of why M&A remains about more than growth. Sometimes companies acquire competitors because the existing structure of an industry no longer works. The media industry has spent years building separate streaming services, only to discover that consumers do not want to pay for every individual service. The logical response is consolidation. Whether combining two complicated media businesses actually creates the promised efficiencies is another matter.
The deal is also a useful reminder of the increasing regulatory cost of very large transactions. Although the US Department of Justice has cleared the merger, a coalition of 12 states and the Writers Guild of America continues to challenge it on antitrust grounds. The transaction includes a ticking fee for WBD shareholders if closing is delayed, illustrating how much economic pressure can accumulate while regulators and courts consider a megadeal.
Further down the technology list is SpaceX's $60 billion all-stock acquisition of Anysphere, the company behind AI coding platform Cursor. The deal is another example of how quickly AI applications are becoming strategic assets. Cursor had reportedly reached around $2.6 billion in annualised enterprise revenue despite being a relatively young company, and SpaceX intends to combine its developer capabilities with Grok and its wider AI operations.
It is a remarkable valuation for a coding company, but it also illustrates an important shift in M&A. Large technology companies are increasingly buying distribution and users rather than simply buying technology. The winners of the AI race may not necessarily be the companies with the best underlying model. They may be the companies that can put those models directly into the workflows of millions of developers, businesses and consumers. Cursor gives SpaceX an established route into software development at exactly the moment AI coding is becoming one of the most valuable commercial applications of generative AI.
Another significant transaction is the $33.4 billion acquisition of AES by a consortium led by BlackRock's Global Infrastructure Partners and EQT, including debt. The equity purchase was around $10.7 billion, meaning the headline figure needs some context, but the strategic rationale is very clear. AES owns power generation and infrastructure at a time when electricity demand is rising sharply because of data centres and AI. Paul Hastings advised Goldman Sachs and Citi on the committed financing for the acquisition.
This is perhaps the strongest example of a broader trend running through 2026's largest deals: investors are increasingly buying the infrastructure surrounding AI rather than AI itself. Data centres need electricity. Electricity requires generation and transmission infrastructure. Companies building AI applications need chips, networks, storage and data. The resulting investment opportunity spreads far beyond companies whose names contain the letters "AI".
The $22 billion acquisition of Roku by Fox is a different kind of strategic bet. Weil advised Fox while Goodwin advised Roku. Fox is paying approximately $22 billion in enterprise value to combine its sports, news and entertainment content with Roku's connected-TV platform, The Roku Channel and more than 100 million global streaming households.
The deal is particularly interesting because it is essentially about owning the relationship with the viewer. Fox has content, but Roku controls an important part of how viewers access and discover content. By combining the two, Fox gains access to Roku's advertising technology and first-party data while giving its own content a much larger distribution platform. In an industry where traditional television audiences are shrinking and advertising is moving online, that strategic logic is difficult to ignore.
Britain has also been a major target for M&A this year, although the story is somewhat different. EQT's takeover of Intertek, agreed in June, values the FTSE 100 testing and certification company at approximately £9.5 billion in equity value and £10.9 billion including debt. Freshfields advised EQT, while Slaughter and May advised Intertek.
The transaction is significant not because Intertek is an obvious AI play, but because it demonstrates another important 2026 trend: international and private-equity buyers are looking at UK-listed companies as relatively attractive assets. UK public companies have often traded at discounts to comparable businesses in the US, making them potential takeover targets. In the first half of 2026, overseas bidders accounted for 94% of UK public M&A deal value, with US bidders responsible for £14.9 billion.
That makes the UK market particularly interesting from a legal perspective. Public takeovers require specialist knowledge of the Takeover Code, shareholder engagement, financing, competition law and increasingly complex political and regulatory considerations. The Intertek deal also shows how the traditional private-equity playbook is evolving: rather than simply buying smaller private companies, sponsors are increasingly willing to take major public companies private when they believe the public market is undervaluing them.
The pattern becomes even clearer when looking at Zurich's £8.2 billion acquisition of Beazley. Slaughter and May advised Zurich, while Freshfields advised Beazley. The strategic logic was to combine Zurich's global scale with Beazley's specialist insurance capabilities and access to Lloyd's, creating a specialty insurance group with approximately $15 billion in gross written premiums. Zurich expected around $150 million of annual pre-tax cost savings by 2029.
Beazley is an interesting deal because it looks almost boring compared with SpaceX and xAI, and that is precisely why it matters. Not every major transaction in 2026 is a bet on a futuristic technology. Established companies are still pursuing conventional consolidation where there is a clear strategic fit, particularly in industries where scale, specialised expertise and regulatory access create competitive advantages.
Taken together, these transactions suggest that 2026's M&A boom is being driven by several forces at once. AI is undoubtedly one of them, but the story is larger than a rush to buy artificial intelligence companies. Companies are acquiring the electricity, computing infrastructure, distribution networks and industrial assets that AI requires. At the same time, traditional industries are consolidating to cope with changing consumer behaviour, competitive pressure and inefficient capital markets.
Private equity is playing a major role as well. EQT's Intertek deal and the AES transaction demonstrate that sponsors still have enormous amounts of capital to deploy, while London's relatively low valuations have made UK-listed companies particularly attractive targets. UK public companies received more than $231 billion of takeover interest by the middle of the year, more than double the equivalent level in 2025.
Perhaps the biggest change, however, is the scale at which businesses are now willing to act. The first half of 2026 saw six deals worth more than $50 billion and 47 deals above $10 billion. That suggests boards are becoming more willing to pursue transactions that would have looked almost unthinkably large only a few years ago.
But size does not equal success. The logic behind many of these deals is compelling on paper: synergies, scale, infrastructure, strategic positioning and access to new markets. The challenge is turning that logic into reality. Integrating two large businesses is difficult, particularly when they operate under different regulatory regimes, corporate cultures or technology systems. The history of M&A is full of transactions where the strategic rationale looked perfect but the promised benefits failed to materialise.
That is what makes 2026's M&A market so interesting. It is not simply a return of the megadeal. It is a market in which companies are trying to position themselves for a rapidly changing economy. AI is accelerating the demand for power and infrastructure, streaming is forcing media consolidation, private equity is taking advantage of undervalued public companies and insurers are seeking scale in increasingly specialised markets.
The biggest deals of the year therefore tell us something more useful than which companies are willing to spend the most money. They show where corporate executives, private-equity investors and their advisers believe the world is heading. The real test will come later: whether the enormous premiums and billions in promised synergies can actually produce the returns that made these deals seem worthwhile in the first place.