By Edward Pitsi, Co-Founder and CEO
Private equity (PE) is entering a phase where artificial intelligence (AI) must become an embedded capability across the investment lifecycle rather than a peripheral consideration. The pace of adoption is accelerating, and early movers are already beginning to see meaningful advantages in speed, insight generation, and operational capability. At the same time, expectations are shifting. Global investor conversations point to a growing expectation from LPs for GPs to have a perspective on how they will utilise AI to enhance their offering.
Yet, despite the momentum, the industry remains in what can best be described as a learning and experimentation phase. This is a moment for curiosity, structured experimentation, and disciplined learning.
There is no established AI playbook among leading global peers, and this is even more pronounced in emerging markets, where AI adoption within private equity remains largely uncharted territory. South Africa is a particularly clear example: engagement tends to stay at the conceptual level, with discussions often centred on generic generative AI use cases rather than the specific, operational realities of PE.
AI across the investment lifecycle
AI is best understood as a horizontal capability that cuts across every function within the GP ecosystem, from origination through to exit and fundraising, as well as back office and support services. Its value compounds when applied holistically rather than being isolated to one stage of the investment process.
In sourcing, AI is already enhancing the ability to identify and prioritise potential targets more efficiently. Effectiveness remains heavily dependent on data quality and availability, which is particularly relevant in South Africa. Global examples such as Hg illustrate how AI can be used to mine large company datasets to identify merger and acquisition targets, and sales prospects with specific characteristics, effectively strengthening origination through pattern recognition and data-driven targeting. This opens the door to more sophisticated deal pipeline tools that flag investment fit and prompt internal action while surfacing broader market patterns that improve proactive sourcing strategy.
In diligence, AI is proving particularly powerful in the rapid summarisation of large and complex datasets. However, it remains limited in its ability to provide context, nuance, and access to private or informal information. Its role is therefore to supplement human judgement. One of its most immediate applications is in accelerating the review of legal documentation and dense diligence materials across deal teams.
In value creation, the opportunity becomes even more compelling. AI enables pattern recognition across portfolio companies, supporting operational efficiencies and revenue enhancement. Critically, it should be viewed as a lever to move earnings and margins, rather than simply optimise processes. This creates the potential for repeatable AI playbooks that can be deployed post-acquisition, as well as the ability to analyse investment reports and quarterly packs to identify underperformance risks or signals for optimal exit timing.
Across reporting and monitoring, AI enables automated performance tracking and early risk detection. It also supports standardisation across portfolios and enhances meeting intelligence, allowing firms to track LP conversations over time and to surface recurring themes or misalignment in messaging.
In fundraising, AI is already improving speed and responsiveness in LP due diligence processes, while also enhancing communication and transparency. Over time, AI-enabled systems may also be able to extract LP preferences from historical interactions or publicly available information and tailor future engagement more precisely.
Operationally, there is growing interest in AI-native fund infrastructure. Hanover Park has rebuilt its fund operations from the ground up. Its platform integrates fund accounting, portfolio management, and investor reporting into a single system, using AI agents to automate workflows such as data extraction, journal entries, and reporting, thus enabling real time visibility and significantly reducing manual effort.
At the same time, firms are actively debating whether to build, buy, or partner for these capabilities, with limited internal technical resources driving interest in alternative models. Some firms are exploring external AI enablement hubs or JV structures with AI providers. Apollo Global Management, for example, has established a Centre of Excellence supported by external AI experts and a broader partner ecosystem, enabling faster access to specialist capability while accelerating implementation across its portfolio.
In research and market intelligence, AI is being used to improve internal knowledge retrieval and map broader private equity ecosystems, helping to identify patterns and strategic opportunities across regions. However, a key constraint remains the inability to operationalise insights without dedicated technical support.
Key issues and concerns
Despite the enthusiasm, several challenges remain. Ethical considerations, including job displacement and data privacy, are increasingly part of internal discussions. More significantly, adoption requires a meaningful cultural shift. Internal AI champions, structured training, and experimentation are essential, yet resistance to change remains a real barrier in many organisations.
A further strategic concern is the potential erosion of information asymmetry, which has historically been a core advantage in PE. Compounding this is the lack of dedicated internal resources focused specifically on AI implementation, resulting in fragmented, often individual-led adoption rather than firmwide integration. In many cases, success will depend on creating a dedicated AI ownership function rather than treating it as an extension of existing roles.
Lessons from global peers
Even among the largest firms, this remains an evolving space, reinforcing that the industry is still experimenting rather than converging on a single model.
While there is no single approach to AI mobilisation in PE, clear patterns are emerging. Some firms are building deep internal capability, while others are creating hybrid models that combine internal teams with external expertise.
EQT’s proprietary AI platform, Motherbrain, which has become deeply embedded across its investment process, uses machine learning to analyse vast datasets on companies, markets, and relationships, helping EQT to identify investment opportunities and map competitive landscapes and surface patterns that traditional sourcing methods would miss. By centralising data and institutional knowledge, Motherbrain enhances both the speed and scale of deal sourcing while also supporting portfolio value creation.
An example of large-scale mobilisation is Vista Equity Partners, which has taken an enterprise-wide approach to embedding generative AI across its portfolio. The firm has built a dedicated internal capability to support more than 85 portfolio companies, helping them apply AI across product development, sales, operations, and research and development. It also requires companies to set measurable AI targets as part of their annual planning and actively tracks progress against these goals. Beyond governance, Vista drives adoption through practical mechanisms such as portfolio-wide hackathons and peer learning forums, ensuring that experimentation translates into real, scalable commercial outcomes.
While there is still no consensus on build versus buy versus partner, competitive advantage is increasingly driven by speed of learning. Human oversight remains critical, with AI acting as an augmentation layer for investment judgement.
Infinite Partners’ outlook
At Infinite Partners, we believe that data constraints remain a limiting factor for South African entities, but there is still a significant opportunity to start with existing tools and unlock early wins.
In our view, the long-term advantage will come from embedding in-house expertise rather than relying on fragmented external solutions, and the focus must remain on high impact use cases aligned to strategy rather than experimentation for its own sake. We think that, ultimately, the biggest opportunity may lie in enabling portfolio companies themselves through AI-driven transformation.
Internally, we are actively building capability and embedding AI into our investment processes. Our approach remains practical, focused on value creation, and we are focusing on the basics first. Our emphasis is on structured learning and disciplined experimentation.
As LP scrutiny of AI adoption continues to increase, the expectation is shifting from awareness to enablement. We think that the firms that succeed will be those that commit early, build capability, and act decisively. The firms pulling ahead are not those with all the answers. They are the ones learning fastest, experimenting openly, and steadily embedding AI into how they create value.
One thing, however, is certain: in time, AI will become embedded across every stage of the PE investment lifecycle, and the ability to engage with it thoughtfully will emerge as a defining differentiator between firms. Yet even as the technology matures, relationships and trusted networks will remain at the heart of the industry. Deals are still won on conviction, judgment, and the credibility built over years of working alongside management teams, co-investors, and advisors. Algorithms can sharpen analysis and accelerate diligence, but they cannot replicate the trust that underpins a handshake, the read of a room, or the instinct that comes from deep human relationships. PE, for all its sophistication, remains above all a people business.
