Investment firms are heading into 2026 with AI budgets largely committed, but the industry’s enthusiasm is being tempered by a familiar set of obstacles: messy data, weak governance and lingering uncertainty over what the technology can safely do.
ScienceSoft’s Q4 2025 Investment AI Market Watch suggests the sector has moved past the stage of asking whether generative AI matters. The more urgent question is where it can deliver value without exposing firms to compliance, pr...
Continue Reading This Article
Enjoy this article as well as all of our content, including reports, news, tips and more.
By registering or signing into your SRM Today account, you agree to SRM Today's Terms of Use and consent to the processing of your personal information as described in our Privacy Policy.
ivacy or reputational risk. Wealth managers are already earmarking a meaningful share of technology spending for AI, and a broad majority of firms now treat it as a strategic priority. Yet much of the market still appears stuck between experimentation and scaled deployment.
That tension is reflected in the wider technology landscape. Gartner said in January that worldwide spending on AI is set to reach $2.52 trillion in 2026, driven largely by infrastructure investment, while KPMG’s latest AI Pulse survey found that most business leaders intend to keep funding AI even if economic conditions soften. At the same time, KPMG also reported that the complexity of agentic systems remains one of the biggest barriers to adoption.
For investment firms, generative AI has become the first technology to break through from pilot projects into practical use. The most common applications are now relatively narrow and controlled: research summarisation, meeting preparation, drafting client communications and internal decision support. These are exactly the areas where the technology can save time without being asked to make unsupervised judgments. Regulators have also helped define those boundaries. FINRA has pointed to data extraction, summarisation, conversational assistance and content drafting as the most common uses among member firms.
Vendors have responded by tailoring products to those needs. BlackRock has extended its Aladdin platform with generative reporting features, while other wealthtech providers have introduced tools aimed at search, note-taking, workflow support and communication drafting. The pattern is clear: the industry wants AI that helps professionals work faster, not systems that try to replace them.
Agentic AI, by contrast, remains earlier in its journey. Unlike copilots that respond to prompts, agentic systems can carry out multi-step tasks across different platforms with limited intervention. That makes them attractive for client servicing, onboarding, portfolio workflows and operations, but it also makes them harder to govern. ScienceSoft said only a small share of investment firms were using agentic AI by the end of 2025, even though interest is rising as firms look for ways to ease staffing pressures and automate repetitive work.
The cautious pace is not surprising. Cisco’s 2026 Data and Privacy Benchmark Study found that most companies have expanded privacy programmes because of AI, and a large majority plan further investment to cope with the complexity. It also reported that many organisations still struggle to access high-quality data efficiently. For investment firms, that is a particularly acute problem, because their systems are often fragmented across custodians, CRM platforms, research tools, compliance systems and legacy portfolio technology.
That data problem is helping to slow the move from experimentation to scale. ScienceSoft said many firms are still not ready to move beyond pilots, and that low trust in outputs, privacy concerns and unclear regulatory expectations continue to suppress adoption. The consultancy also argued that firms increasingly understand that modern AI depends less on model novelty than on the quality of the infrastructure underneath it.
There are signs that investors in the sector are becoming more selective as well. S&P Global has noted a broader shift in AI funding from indiscriminate enthusiasm towards more disciplined scrutiny of revenue, customer return and infrastructure strength. In wealth management, that is translating into a preference for practical, explainable use cases over speculative automation. Morgan Stanley has also argued that AI-related capital spending is changing financing behaviour across markets, with infrastructure-heavy projects drawing more attention from the credit side.
The likely result is that 2026 will be less about sweeping AI transformation than about foundation-building. Firms will keep funding AI, but many will spend just as much on governance, integration, security and data readiness as on the models themselves. In investment management, that may prove to be the more important bet.
Source: Noah Wire Services
