Artificial Intelligence · Profile from AI Research Companion
Infosys
The Bengaluru IT-services company, founded in 1981 by N. R. Narayana Murthy and six colleagues, that helped build India's offshore outsourcing industry and is now one of the clearest test cases of what generative AI does to labour-priced services. Infosys earns ~$20bn a year (FY2026, company) mainly by maintaining, modernising and running software and business processes for large Western companies, using ~323,000 staff (March 2025, verify). Its AI strategy has two eras. Under Vishal Sikka (2014–17) it tried to become an AI-platform company with Mana and Nia, an effort that ended with his resignation amid a governance fight with the founders. Under Salil Parekh (CEO since 2018) it launched the Topaz AI services brand (May 2023), reported AI revenue rising to 10.5% of the total by the September 2026 quarter (company), and won large deals partly by promising to pass AI productivity on to clients. Total growth, however, slowed to ~1–3% a year, guidance was cut twice in 2026, and its shares had their biggest one-day fall in years when Anthropic released Claude Cowork plug-ins on 4 February 2026 (reported). This profile covers Infosys's AI strategy and AI exposure, not its full history.
At a glance
- Founded
- 2 July 1981, Pune, by N. R. Narayana Murthy, Nandan Nilekani, S. Gopalakrishnan, S. D. Shibulal, K. Dinesh, N. S. Raghavan and Ashok Arora; moved to Bengaluru in 1983
- Headquarters
- Bengaluru, India; large campuses in Mysuru (training), Pune, Hyderabad, Chennai and elsewhere; delivery and proximity centres in the US, Europe and Australia
- Ownership / structure
- Public: listed in India (NSE/BSE: INFY) since 1993 and in the US via ADRs (Nasdaq 1999, NYSE from 2012); founders and families hold a minority stake (~14–15%, verify)
- Leadership (2026)
- Salil Parekh (CEO and managing director since January 2018); Nandan Nilekani (co-founder, non-executive chairman since August 2017); Jayesh Sanghrajka (CFO, verify)
- AI products and services
- Topaz (AI-first services, solutions and platforms, launched May 2023); Topaz Fabric (model-agnostic AI services stack, 2025–26, verify launch date); agentic AI offerings (an "Agentic AI Foundry" within Topaz, verify); Finacle banking software and EdgeVerve platforms with AI features; earlier Mana (2016) and Nia (2017) platforms
- Scale
- Revenue >$20bn in FY2026 (to 31 March 2026; +3.1% in constant currency); Q2 FY2027 revenue $5,145m (+1.5% year on year in constant currency); large-deal TCV $14.9bn (FY2026) and $8.8bn in H1 FY2027 (company); AI revenue 5.5% of total (Q3 FY2026), 8.2% (Q1 FY2027) and 10.5% (Q2 FY2027) (company)
- AI exposure
- Most revenue comes from application development, maintenance, testing, infrastructure and business-process work billed by effort — the categories coding and service agents most directly automate
- Key partnerships
- Hyperscalers (Microsoft, Google Cloud, AWS); Nvidia (2023 partnership to train staff and build generative AI solutions, verify); software partners (SAP, Salesforce, ServiceNow, Oracle); OpenAI pledge (December 2015)
- Signature moves (incl. failures)
- Mana, Nia and the Sikka AI-first era (2015–17, failure); Topaz (2023); the $2bn AI-and-automation deal and productivity pricing (2023); the cancelled $1.5bn AI-platform MoU (2023, failure); the Mysuru trainee terminations (2025, controversy); the record buyback and the H-1B fee (2025); the agentic AI push and Topaz Fabric (2025–26); the Anthropic plug-in sell-off and AI-revenue disclosure (2026); the 2026 guidance cuts (failure)
9 strategic moves analysed
Each one gets a full strategy file in the companion: context, the decision, the financing, what happened, and the lesson.
2016
Mana Nia and the Sikka AI First Era
Launch / leadership change (failure)
2023
Cancelled 1 5bn AI Platform MoU
Contract (failure)
2023
Topaz AI First Services Launch
Launch
2023
Two Billion AI Automation Deal and Productivity Pricing
Contract / pricing
2025
Agentic AI Foundry and Topaz Fabric
Launch
2025
Mysuru Trainee Terminations
Restructuring (controversy)
2025
Record Buyback and the H1B Fee
Capital return / regulatory
2026
Anthropic Plugin Selloff and AI Revenue Disclosure
Market event / disclosure
2026
Guidance Cuts and the Low Growth Squeeze
Guidance (failure)