Artificial Intelligence Integration Enhances Appraisal Process, Human Oversight Remains Mandatory

Artificial Intelligence Revolutionizes Performance Management in India's IT Sector

In a significant shift, performance management in India's IT and services sector is undergoing a transformation from an annual "autopsy" to a data-driven process, with artificial intelligence (AI) playing a crucial role. At IT firm HCLTech, AI now participates across quarterly and annual check-ins, providing context-rich review notes to managers without independently influencing outcomes.

According to a company spokesperson, the AI system is designed to be advisory, equipping managers with insights that help them make more informed decisions. However, managers remain "firmly accountable for final evaluations." Similarly, IT services firm Happiest Minds Technologies is using AI to assess the quality of feedback, consolidate insights, and simulate performance conversations. However, the system stops short of generating feedback or assigning ratings, with the company's Senior Director, People Practice, Puja Rungta Agarwal, stating that accountability and ownership "remain firmly with managers."

Read also: India Records First Weekly Inflows in Seven Weeks Amid Ongoing Global Liquidity Support

The Boundary Between Assistance and Authority

The key to successful AI implementation lies in striking a balance between assistance and authority. IT & BPM services company Digitide Solutions said AI handles the mechanics, while humans handle the meaning during the appraisal season. The company's systems automate data aggregation and trend analysis, but stop at recommendations. "Final decisions on ratings, promotions, compensation, and career progression remain 100 percent human-led," said Digitide CEO Gurmeet Chahal.

The Bias Paradox

One of the strongest arguments for AI in appraisals is its hypothetical objectivity. Research from the University of New Hampshire found that employees tend to trust AI evaluations more when they expect bias from human supervisors, viewing algorithms as less influenced by personal preferences. However, the paradox is that bias shifts and doesn't vanish. Workforce management force Hunar said that most bias in performance evaluation is introduced by humans, by fatigue, familiarity, and pattern-matching, which AI can help surface but not fully eliminate.

Read also: Navigating Unpredictable Markets: A Guide to Intraday Trading Strategies

Data Quality: A Growing Concern

The growing reliance on AI also brings a more basic challenge, i.e., data quality. "Employees need to know when AI is being used, what information it is looking at and how it is arriving at suggestions," said Aditya Narayan Mishra, Managing Director and CEO of CIEL HR. "If employee records or performance inputs are incomplete or inaccurate, the output will be flawed." This concern is backed by academic work, which highlights the importance of data completeness and reliability in AI-led appraisal systems.

Rewriting the Manager's Role

What is changing most is not the rating itself, but the role of the manager. Earlier, performance reviews depended heavily on memory, documentation, and narrative building. AI consolidates multiple signals into clear, actionable insights, said Gurmeet Chahal, likening it to telemetry in high-performance racing where data informs but does not decide. Moreover, Happiest Minds' simulation tools are even being used to train managers on how to conduct performance conversations.

The Line That Won't Be Crossed

The stakes are too high, involving pay, promotions, and careers. And while AI can process scale and detect patterns, it still lacks context such as personal circumstances, team dynamics, or one-off events that often shape performance. "Automation compresses time and removes friction... But telling a manager which employee to promote... that's a recommendation. Never a decision," said AI platform Hunar.