After the valuation crash of software stocks, is the era of AI "mega mergers and acquisitions" here?

Deutsche Bank believes that most companies are currently far behind market expectations in AI implementation, and the volatility in AI-related market capitalization is prompting firms to accelerate mergers and acquisitions.

According to Wind Trading Desk, on February 26, Deutsche Bank’s research team published a report stating that recent stock market fluctuations and the sell-off of AI concept stocks have forced CEOs to quickly develop AI strategies and clearly communicate these strategies to investors.

However, by 2025, only 11% of companies may have fully implemented at least one AI-related business function. This means that most CEOs face significant pressure to accelerate AI adoption. In response to the implementation pressure, M&A is becoming a key means for many CEOs to catch up with their peers.

Data shows that the scale of external transactions involving private AI companies (including acquisitions, minority investments, private placements, and public offerings) has skyrocketed from negligible levels before 2013 to nearly $40 billion annually between 2021 and 2024.

(Scale of global external transactions involving private AI companies)

The report suggests that the valuation revaluation of the software sector is at an all-time high, private AI M&A activity continues to heat up, and the pace of global M&A is becoming more segmented. These three themes will profoundly influence asset allocation decisions over the next one to two years. Meanwhile, regulatory uncertainty and regional differences in macro interest rate environments will be the biggest variables affecting M&A pace and pricing.

Most companies lag seriously in AI deployment, and CEOs are under immense pressure

Deutsche Bank points out that AI adoption is currently uneven, with startups and large corporations as early adopters. The report cites data from a survey in Q2 2025:

• Only 8% of companies plan to fully deploy at least one AI function by mid-year;
• Only 3% expect to do so by year’s end;
• 11% explicitly state they have no plans to implement intelligent agent AI.

The International Monetary Fund (IMF) estimates that about 40% of jobs worldwide will be affected by AI, especially “cognitive” work. Analyzing keywords from S&P 500 earnings call transcripts shows:

• AI and machine learning remain the top hot topics, with layoffs, chip shortages, and R&D investment also among the fastest-growing topics;
• Discussions related to M&A have rebounded significantly after the tariff shock in spring 2025, with mentions surpassing dividends and buybacks;
• The fastest-growing capital allocation theme in the past six months has been capital expenditure and R&D.

**** (Increase in mentions of specific topics during S&P 500 earnings calls)

Looking at individual companies, industry leaders like Marriott International, Amgen, and S&P Global have explicitly expressed positive strategic attitudes toward AI in their earnings reports, viewing it as a net business benefit rather than a threat.

Notably, mid-sized companies with 50-249 employees show significantly lower AI usage rates.

They lack the agility and focus of startups, and do not have the resources and data scale of giants, making them most likely to fall behind in the race. Acquiring ready-made AI capabilities through M&A is a practical shortcut for them.

Software valuations plummet, and M&A windows quietly open

Fortunately, the market has provided an acquisition window.

Since the market peaked in mid-January, the software and services sector has been the worst-performing industry group in the Russell 1000 index, with a median decline of 25%. Its valuation ranking dropped from third to ninth place.

(Performance of the software sector in Russell 1000 since January 12)

More importantly, when adjusting for growth expectations, the valuations of software companies have become relatively average. In the US market, the PEG ratio ranking dropped sharply from 7th to 17th; in Europe, from 3rd to 15th. The valuation bubble has been significantly squeezed, giving corporate buyers more bargaining power at the negotiation table.

(PEG ratio ranking based on growth expectations, dropping from 7th to 17th)

Looking ahead, M&A activity in the US is expected to remain cautious, while Europe shows a “hot and cold” pattern. Deutsche Bank’s M&A leading indicator indicates:

US: The rebound in M&A activity in Q1 may slow down in Q2 due to rising policy uncertainty and mixed signals from capital issuance;

(M&A momentum in Q2 2026 may weaken)

Eurozone: Rising sovereign bond yields are weighing on M&A prospects, with short-term pressure;

UK: Benefiting from lower bond yields and strong stock market performance, M&A recovery is expected to accelerate faster than current market expectations.

(Forecast for M&A deal volume in Eurozone and UK over the next three months)

What kind of AI companies are most likely to be acquired? Deutsche Bank believes that the more specialized an AI company is, the more attractive it is to industry giants. They need tools that delve into specific verticals and solve particular problems.

Private equity-led deals, but exits are inevitable

A key structural change in the market is the surge in the share of private equity and other financial buyers in global software M&A transactions.

Data shows that the share of private equity and similar financial buyers has soared from 28% in the 2000s to 72% in the 2020s, while non-tech corporate software M&A has shrunk from 17% to 5%.

(Software M&A by deal size and buyer type globally)

These large private equity deals will eventually need to be exited. Selling assets to entities seeking AI capabilities will be a key exit route.

The report cites data indicating that from 2022 to 2024, M&A transactions accounted for an average of 42% of external deals involving private AI companies, while IPOs only accounted for 3%.

Many AI challengers are small and continue to lose money, while large incumbent firms possess proprietary data, trust, and scale advantages—especially in highly complex, regulated industries where startups can hardly replicate.

Risks and lessons from history

M&A is not a panacea. Failures in integration, cultural clashes, loss of key talent, and high ongoing costs are risks.

Deutsche Bank notes that the number of AI-related bills proposed in the US Congress surged from about 80 in 2022 to over 200 in 2024, increasing regulatory uncertainty.

(US Congress AI-related bill proposals growth)

History offers a long-term perspective. During the tech boom of the 1990s, the Nasdaq experienced multiple declines of over 10%, but the average annual return still reached 32%.

Regulatory evolution at that time ultimately strengthened scale effects, leading to market concentration. This time, giants with capital, data, and scale advantages may similarly secure a more favorable position in the long AI race.

The report suggests that what makes this moment unique is that, during the rise of the AI wave, large tech companies are enjoying unusually abundant free cash flow. They are among the few entities worldwide capable of bearing massive AI capital expenditures and potential losses. The entry barrier for this race has been high from the start.

Ultimately, for investors, the AI M&A cycle is shifting from conceptual to substantive implementation. Valuation resets create potential strategic buying opportunities, but regulatory risks, opaque private company valuations, and macro uncertainties remain major constraints. In the medium term, companies capable of actively managing AI M&A strategies will have a competitive edge in reshaping the landscape.

This insightful content is brought to you by Wind Trading Desk.

For more detailed analysis, real-time insights, and frontline research, please join【**Wind Trading Desk ▪ Annual Membership**】

Risk warning and disclaimer

Market risks are present; investment should be cautious. This article does not constitute personal investment advice and does not consider individual users’ specific investment goals, financial situations, or needs. Users should consider whether any opinions, viewpoints, or conclusions herein are suitable for their particular circumstances. Invest accordingly at your own risk.