By Arun Patel, Founder and Partner at Arunasset Investment Services
Artificial intelligence is emerging as a powerful productivity tool for India’s wealth management industry. AI perceivably enables advisors to accomplish in minutes what previously required hours of research. However – the to ascertain the accuracy of the reports requires a subject matter experts. And greater efficiency should not be confused with better investment decisions. These two points cannot be overstated.
To understand its limitations, we must first understand large language models (LLMs), which power ChatGPT and are increasingly integrated into enterprise software, including research platforms, client-management systems and knowledge-management tools. They are probabilistic models that generate language from learned patterns. This means they are a tool that gives you language that the model believes the user will accept. It does not make them reliable forecasters or even at times data gathering tools. An answer can sound convincing yet contain errors, outdated information or questionable assumptions.
Perhaps the biggest immediate advantage is the speed at which information can be retrieved and processed. With the caveat that accuracy is not the main goal, the perception of accuracy is. An advisor can compare annual reports, extract trends from quarterly results, examine management commentary and screen thousands of securities much faster than before. AI can also track changes in valuations, portfolio exposures and financial ratios. This reduces the time spent gathering data and creates more time for interpreting it. Faster access to information, however, is not the same as better investment judgement. And a keen eye for accuracy from the Human side is very important.
AI also creates opportunities for more personalised portfolios. With appropriate, secure client data, models can analyse asset allocation, liquidity needs, investment horizons, tax considerations and exposure to different risks. They can simulate how a portfolio might behave under rising interest rates, an earnings slowdown or an equity correction. Yet a family’s changing obligations, business circumstances or ability to withstand losses are not always captured by a questionnaire. Personalisation requires understanding the person, not merely the portfolio.
Understanding market cycles remains fundamental. Models trained on historical relationships can struggle when those relationships break down. A strategy that worked during abundant liquidity and falling interest rates may behave differently in an inflationary or credit-constrained environment.
Similarly, identifying an attractive stock involves more than comparing ratios. Assessing management quality, competitive advantage, corporate governance, future growth and the price worth paying requires context, scepticism and human judgement.
From an investor’s perspective, human behaviour may be even more important than analytical sophistication. Fear encourages selling after sharp corrections; greed encourages buying after extraordinary rallies. During volatility, an experienced advisor does more than present probabilities. They explain what has changed, distinguish temporary price movements from deteriorating fundamentals and help clients avoid decisions that undermine long-term goals. Trust matters a lot especially when markets are falling and confidence is low.
The golden principles of investing have not changed. And that cannot be stressed enough. Increased computing power does not provide a cheat code. Invest with a long-term horizon, diversify sensibly, maintain an asset allocation appropriate to your needs, insist on reasonable valuations and allow compounding the time to work. Understand what you own, maintain adequate liquidity and avoid excessive leverage. Above all, recognise that a wonderful business is not automatically a wonderful investment at any price. Neither the most sophisticated algorithm nor the newest technology can repeal the relationship between risk, valuation and return.
That perspective matters amid the excitement surrounding AI itself. Technological revolutions create genuine opportunities, but enthusiasm can also lead investors to forget basic principles. Sir John Templeton famously warned that the four most dangerous words in investing are, “This time it’s different.” With AI dominating the investment narrative, the warning is as relevant as ever.
There are also questions of accountability and confidentiality. SEBI’s rules make regulated intermediaries responsible for the security of investor data and for outputs they rely on from AI tools. Human verification, transparent processes and clear responsibility are therefore necessities.
The future of Indian wealth management is likely to belong to advisors who use the computing power of AI, while retaining independent judgement, an understanding of cycles and the trust of their clients. AI may throw computing power behind how decisions are researched but enduring investment wisdom must still determine how they are made.