The AI Imperative: redefining value creation across consulting, investment, and private equity

13 min. read | Download PDF here

Executive summary

Artificial Intelligence (AI) must now be a strategic operating system for investment firms; it is no longer an experimental technology. The firms that institutionalize intelligence into their decision-making, operating structures, and deal workflows will define the next era of alpha generation and advisory value, yet many are still asking the wrong questions or, worse still, asking the right ones too late.

In this conversation with Rajesh Krishnamachari, former Head of Investment Insights at GIC (and member of the Corporate Management Committee) and one of the foremost thinkers at the intersection of finance and AI, we explore how investment managers, private equity operators, and advisory firms can rewire their models around AI. From automating diligence to scaling LLM-driven insights to codifying institutional memory, AI is the key operating mandate of the future.

You can’t lead AI from PowerPoint. You need talent fluent in business, technology, and execution.
— Rajesh Krishnamachari

The state of AI: Investment vs. hype

Understanding that we have collectively reached a point of exhausted saturation around AI hype, separating signal from noise has never been more critical. According to Krishnamachari, AI's impact on investment and advisory spans a wide spectrum of maturity depending on the use case and market segment; in other words, AI is not monolithic. He categorizes AI deployment across two primary vectors:

• Quantitative AI Techniques: rooted in mathematical rigor, already embedded in alpha generation, execution, and risk.
• Generative AI & LLMs: rapidly evolving but still early-stage in their enterprise reliability and auditability. 

We must judge each case not just on technological sophistication but on enterprise readiness, data availability, and process hygiene. In the chart below, we take a brief look into several investment functions and their readiness for AI supplementation.

AI readiness landscape in investment functions

Function	AI readiness	Description
Derivative hedging	Enterprise-ready	Reinforcement learning models are actively optimizing hedge ratios
Smart order routing	Mature	ML is driving more efficient trade execution across fragmented venues.
Portfolio analysis	Scalable	AI enables multi-scenario stress testing, exposure analysis, and heatmapping.
Autonomous signal discovery	Research-phase	Still largely academic; lacks transparency, explainability, and reliability.
Robo-advisory	Nascent	Limited adoption in institutional settings due to trust, compliance, and UX constraints.
LLM portfolio co-pilots	Early-stage	High interest for research augmentation; adoption growing, but governance immature.

Public markets have more data, but private markets have better process hygiene: that makes them more AI-ready.
— Rajesh Krishnamachari

This observation cuts to the core of AI’s applicability: public markets may offer rich data signals, but their fluid, often unstructured workflows hinder AI orchestration at scale. In contrast, private equity and real estate operate with templated workflows, repeatable diligence protocols, and central deal repositories, making them ripe for LLM deployment and rule-based automation.

Hype cycle meets operational reality

While some firms race to build GPT-powered investment co-pilots, Krishnamachari emphasizes that a clear gap exists between innovation theater and operational AI. The most mature firms are already:

• Replacing traditional rule-based macros with reinforcement learning agents
• Automating signal validation with ensemble models
• Enriching analyst workflows with retrieval-augmented generation (RAG) systems
• Embedding AI in backend processes like fund accounting and LP reporting

Fully autonomous portfolio construction, generative investment memos without human oversight, and AI-powered deal sourcing devoid of human validation are all still in need of refinement before firms can deploy them. These actions, however, present a promising horizon for future supplementation and implementation of AI.

AI doesn’t yet replace judgment—it amplifies it. The firms winning today are focused on augmenting human edge, not replacing it.
— Rajesh Krishnamachari

From pilot projects to core systems

Investment leaders must evaluate every AI initiative through three lenses:

1. Data governance – Do we have structured, high-quality data ready for AI training and inference?
2. Business alignment – Does the AI initiative map to a profit lever or strategic KPI?
3. Talent readiness – Do we have domain experts who can interpret and deploy model outputs?

When those three intersect, firms move from pilot purgatory to true transformation.

The biggest risk isn’t AI replacing your analysts. It’s your competitors using AI to outpace your analysts.

Boardroom framework: A dual-track strategy for AI deployment

Most executive teams treat AI as a technology investment. But as Krishnamachari makes clear, AI is a business model reinvention strategy not an IT initiative.

You must walk backward from your business model. Then forward into what’s possible. Where they intersect, that’s where AI becomes real.
— Rajesh Krishnamachari

AI adoption must follow two complementary tracks:

The dual-track AI deployment framework

Walk backward	Walk forward
Start with strategic objectives and business model	Begin with experimentation and use cases
Ask: “What must be true for this to drive value?”	Ask: “What can we try today with current tools?”
Define P&L-level outcomes and talent needs	Explore prototypes, fast feedback loops
Translate business goals into AI architecture	Evolve into scalable, repeatable solutions

Where the tracks intersect: Strategic AI activation

Success doesn’t come from walking only one path.

• Too much backward walking leads to AI paralysis; months of strategy without delivery.
• Too much forward walking leads to AI theater; flashy demos with no business impact.

When firms synchronize these tracks—when top-down strategic clarity meets bottom-up feasibility—AI becomes more than an experiment. It becomes embedded in the firm’s DNA.

The best firms are building executive coalitions that combine vision with execution power. That’s when AI stops being a sideshow and starts driving real change.

Four strategic value pillars for AI in investment and advisory

Krishnamachari distills the board-level business case for AI into four foundational outcomes:

AI value pillar	Impact
1. Better performance	Alpha generation via faster data absorption, nonlinear insights, and risk calibration
2. Faster execution	Shorter diligence timelines, streamlined fund ops, and reduced latency across the stack
3. Scalable operations	Leverage AI to amplify analyst output, standardize processes, and reduce headcount strain
4. Institutional memory	Codify decision rationale, investment memos, and best practices into reusable knowledge systems

These pillars do more than justify AI, they align it directly to how investment firms compete and scale.

Executive alignment is the hardest part

AI success isn’t blocked by tooling; it’s blocked by misaligned mental models at the top. Krishnamachari emphasizes that every firm needs an executive narrative around AI that answers the following three main questions:

• What is our firm’s differentiated data asset?
• Where do we compete on speed, precision, or repeatability?
• Which decisions today are manual but pattern-based?

Only after answering these questions can the firm prioritize AI investments that truly matter.

If AI isn’t solving a strategic pain point or enhancing a core process—it’s just expensive theater,
— Rajesh Krishnamachari

Leadership buy-in cannot be limited to a one-time keynote. It must be a system of aligned incentives, C-suite ownership, and shared outcomes.

The alpha opportunity: Investment use cases that deliver

At GIC, Krishnamachari led a comprehensive mandate to reimagine how artificial intelligence could drive performance and institutional intelligence across the investment lifecycle. He distilled his transformation approach into five strategic objectives—each reinforcing AI as a force multiplier:

Five mandates for AI in investment

1. Generate alpha—Leverage AI to identify non-obvious signals and inefficiencies
2. Enable scalable research—Institutionalize insight generation beyond individual analysts
3. Operationalize workflows—Infuse automation into routine yet critical decision cycles
4. Improve collaboration—Build cross-functional connectivity across teams and geographies
5. Communicate & democratize insights—Make intelligence accessible and actionable for decision-makers

AI as co-pilot is here. AI as autonomous researcher is coming.
— Rajesh Krishnamachari

Proven quant use cases

These use cases are already driving measurable performance outcomes in public and private markets:

Use case	Description
Smart order routing	Machine learning optimizes execution pathways to minimize slippage and latency
Reinforcement learning for derivatives	AI dynamically adjusts hedging strategies based on evolving market data
Macro forecasting	Time series modeling to predict regime shifts and economic inflection points
Real estate cap rate modeling	Multi-variable analysis to improve accuracy in illiquid markets

 

LLM & generative AI use cases

Emerging generative AI applications, while early, are proving high-leverage in cognitive-heavy tasks:

Use case	Application
Due diligence automation	Parsing and synthesizing financial documents and data rooms
Internal research co-pilots	LLMs assist in report generation, summarization, and data interpretation
Portfolio monitoring bots	Continuous scanning of risk signals and investment anomalies
Backtest introspection	Natural language querying and summary of historical research performance

 

GIC outcomes (post-AI deployment)

Outcome	Impact
+10% alpha uplift	Resulting from AI-enhanced signal discovery and execution
+30% analyst productivity	Achieved through augmentation, not replacement
10x Personal efficiency	Rajesh cites a 10x gain in his own productivity due to LLM tooling and automation

Together, these use cases represent a strategic retooling of the traditional investment process — not just a marginal improvement. AI is reshaping how insights are generated, how risk is quantified, and how capital is deployed.

Rewiring the deal lifecycle: AI across PE and VC functions

Private equity and venture capital firms have historically relied on pattern recognition, network intelligence, and operational rigor to generate returns. With AI, those capabilities can be codified, scaled, and systematized—transforming instinct into institutional advantage.

Private equity has process rigor and clean workflows. Ironically, that makes it more AI-implementable than many public market use cases.
— Rajesh Krishnamachari

Krishnamachari outlines how each phase of the deal lifecycle can be infused with AI to unlock compounding value. The key lies in turning every phase into a data-rich, feedback-loop-driven system.

AI rewiring of the deal lifecycle

Phase	AI applications	Strategic advantage
Sourcing	Graph mining, CRM enrichment, social scraping	Proprietary relationship intelligence, faster targeting
Diligence	LLM-based data room parsing, historical decision patterning	Speed, bias detection, better signal-to-noise ratio
Creation	Location planning, pricing optimization, cross-sell engines	Monetizable levers for top- and bottom-line growth
Exit	Risk alerts, performance monitoring, market pulse tracking	Smarter timing, proactive mitigation, stronger narratives

Deeper dive: How each phase changes with AI

Sourcing: Digitizing the rolodex

• Enriching CRMs with graph-based learning creates a “relationship cloud” around every investment professional.
• AI scrapes data from LinkedIn, PitchBook, press releases, and databases to identify high-potential connections.
• Proprietary network maps become a firm’s core intangible asset—its IP.

Diligence: Faster, sharper, smarter

• LLMs parse financial statements, legal docs, and customer data within minutes.
• Pattern recognition reveals past decision biases—helping improve future investment judgment.
• Secondaries especially benefit from rapid turnaround analysis.

Value creation: Operational AI becomes a lever

• For consumer, retail, and logistics companies: AI-driven location intelligence optimizes footprints.
• Cross-sell recommendation engines drive incremental revenue in multi-brand or platform portfolios.
• Internal workflow automation (e.g., AP/AR, customer support) boosts efficiency with minimal lift.

Exit: Timing and narrative, enhanced

• Portfolio monitoring bots flag changes in market conditions, sentiment, and KPIs.
• AI curates dashboards that strengthen investor communication, pitch materials, and exit readiness.

By embedding AI across the deal lifecycle, firms move from manual intuition to systematized intelligence.

Every deal becomes a self-reinforcing data loop, informing the next.

Strategizing better AI implementation

Krishnamachari outlines a comprehensive three-part framework for institutional investors and advisory firms that aspire to become the go-to partners for AI-driven transformation. More than branding, this is about operational credibility, embedded capability, and executional rigor.

1. Strategic talent: Codifying AI inside the org chart

To build trust with boards and CEOs, firms must have leaders who bring both domain authority and technical fluency. AI can no longer live in an isolated innovation lab—it must be embedded into decision-making centers.

Key roles include:

• AI-native Chief Officers (Chief AI Officer, Chief Data Officer) responsible for aligning AI with P&L outcomes, not just experimentation
• Embedded technologists inside each investment team and function (e.g., AI specialists working alongside deal teams, risk, compliance)
• Cross-functional advisory boards to guide AI governance, opportunity triage, and ethical frameworks

The next generation of leadership isn’t just data-literate, they are AI-native.

2. Embedded partnerships: Building AI as an ecosystem

No firm can build frontier AI capability in isolation. The leaders in this space partner intelligently with academic researchers, leading-edge labs, and cloud-scale platforms.

Three Key Pillars of the AI Ecosystem:

Type of partner	Examples	Value creation role
Academia	Stanford AI Lab, Andrew Ng, MIT CSAIL	Talent pipeline, cutting-edge R&D, external credibility
Frontier AI labs	OpenAI, Anthropic, Cohere	LLM fine-tuning, model experimentation, access to early-stage capabilities
Tech platforms	AWS, Snowflake, Databricks, Hugging Face	Infrastructure, scalability, model deployment, data orchestration

“You must be plugged into the ecosystem,” Krishnamachari says. “The winners co-create, not observe.”

Successful firms don’t just buy software—they co-develop it. They don’t wait for trendsa—they shape them through early adoption and ecosystem engagement.

3. Built before broadcast: Institutional credibility through in-house deployment

Before advising clients on AI transformation, firms must have tested, refined, and scaled these capabilities internally.

Execution priorities:

• Document internal use cases—show how AI has improved productivity, decisions, or speed in your own teams
• Develop internal training datasets and deploy models with feedback loops to ensure learning and trust
• Create sector-specific playbooks that consolidate learnings from across engagements (e.g., Private Equity AI Playbook, Public Markets LLM Toolkit, Risk Intelligence Copilots)

This practice-first approach turns firms into learning organizations—where insights compound over time and advisory work is backed by proof, not just pitch decks.

You can’t fake fluency. We’re now seeing firms distinguish themselves by doing the hard work: training models, deploying copilots, iterating use cases, and owning the intellectual property that emerges.

From vision to execution

You can't rely on a slogan or a title; AI success is determined on track record.

Firms that succeed here:

• Treat AI not as a tool, but as a strategic muscle
• Invest in transversal talent and internal case studies
• Build networked partnerships that evolve with the AI frontier
• Systematize learnings into repeatable, referenceable playbooks

This is the AI advisory model of the future: fluent, functional, and field-tested.

Getting started: Three immediate actions for investment firms

While the promise of AI can feel daunting, Krishnamachari emphasizes that the path to value doesn’t begin with moonshots—it begins with plumbing, people, and pilots.

These three initial moves serve as high-leverage accelerators for firms looking to build meaningful AI capabilities—fast, but without chaos.

1. Fix the data plumbing

AI doesn’t start with models. It starts with clean, accessible data. And most investment firms are still sitting on fragmented infrastructure, shadow spreadsheets, and siloed knowledge.

If you can’t locate your unstructured data, you can’t model it.

Immediate priorities:

• Inventory your data estate—identify where structured, semi-structured, and unstructured data lives
• Tag and organize—apply metadata to files, dashboards, and systems to enhance discoverability
• Centralize and normalize—create common data schemas across business units
• Automate ingestion—integrate with external data providers, news streams, and market feeds
• Lay the foundation for AI-readiness with cloud-based data lakes or hybrid architectures

Without this foundation, even the best AI tools will fail to deliver signal—or worse, introduce bias and hallucinations.

2. Recruit cross-functional talent

You don’t need a thousand data scientists. You need a dozen translators who can sit at the intersection of AI, finance, and operations.

Don’t just hire ML engineers. Hire leaders who translate strategy into algorithms.

Key roles to build now:

• AI-native business leaders: operators who understand both unit economics and neural networks
• AI product managers: to prioritize use cases, scope MVPs, and orchestrate across functions
• Tech-integrated investment professionals: especially in sourcing, diligence, and risk functions
• AI advisors and coaches: to guide change management and stakeholder buy-in

The firms that win won’t just have a Chief AI Officer. They’ll have a Chief of Every Function fluent in AI.

3. Run pilots and create playbooks

Skip the grand vision decks. Start with one use case that moves the needle, measure outcomes rigorously, and turn it into a repeatable playbook.

Choose one vertical: real estate diligence, CRM enrichment, signal alerts. Deliver value. Scale it.

How to launch AI pilots to win:

• Select a low-risk, high-impact use case (e.g., LLM-based due diligence, smart sourcing automation)
• Assign a cross-functional tiger team: data, product, business, compliance
• Build the first version fast with agile sprints and open-source components
• Collect structured feedback from users, including skepticism and edge cases
• Document not just outcomes—but processes, constraints, and what you’d do differently

These documented insights become the foundation of institutional AI playbooks, which can then be applied across portfolios or client segments.

As Krishnamachari puts it: “Document the lessons—not just the code.”

Summary

Action	Objective	First step
Fix the data plumbing	Make data AI-ready	Inventory, tag, and centralize core data
Recruit translators	Build cross-functional fluency	Hire AI-native business leaders and PMs
Run and document pilots	Create scalable, credible case studies	Pick one vertical and start documenting now

By focusing on foundational execution instead of future hype, investment firms can go from experimentation to enterprise value—and avoid becoming the next case study of AI theater.

Conclusion: The time for AI native investing is now

The message from Krishnamachari is clear: AI won’t replace investors. But investors who use AI will outperform those who don’t.

The real disruption is not about automation. It’s about augmentation—of research, relationships, decision-making, and value creation. And the firms that succeed in this next era will not merely deploy AI tools, they will rearchitect themselves around intelligence.

From tools to transformation

AI is not a feature. It’s a fundamental shift in how investment decisions are made, risks are assessed, and firms are built. To treat it as a “nice-to-have” is to lose ground to competitors who are already turning AI into alpha.

Firms must now transition from:

• Isolated pilots to Integrated intelligence
• Tactical wins to Strategic roadmaps
• Technical adoption to Organizational fluency

The opportunity isn’t just about being more efficient. It’s about building new sources of competitive advantage.

In 10 years, no one will talk about “AI strategy.” It’ll be embedded in your investment strategy, your operations, your value creation playbook.

What AI native leaders will do next

1. Redefine leadership—hiring operators fluent in both financial rigor and machine intelligence
2. Institutionalize learning—turning every AI project into a repeatable playbook
3. Reframe value creation—viewing every portfolio company as a candidate for AI transformation
4. Set the tone from the top—with board and C-suite alignment around AI-enabled execution

AI native investing is not about futureproofing. It’s about front-running the future.