🧠 Food for thought

Implications, context, and why it matters.

OpenAI partnership adds to Infosys’ own AI work

• Infosys is building its own “small language model,” a compact AI system trained on Infosys and industry data. The goal is to support generative AI applications across sectors for clients 1.
• “Project Maximus” added about 80 basis points, or 0.8 percentage points, to operating margin in Q2 FY25. The internal program was part of a broader efficiency push, alongside currency moves, acquisitions and higher variable pay 1.
• Client spending remains uneven. Discretionary budgets are improving in parts of U.S. financial services, while retail and communications stay focused on cost cuts and efficiency 1.
• Infosys closed 21 large deals with a total contract value of US$2.4 billion in Q2 FY25. It also raised full-year constant-currency revenue growth guidance to 3.75% to 4.5%, helped by stronger execution and a double-digit rise in the pipeline for deals below US$50 million 1.

IT services firms are becoming AI’s route into enterprises

• Complex AI models often need integrators such as Infosys to connect with older corporate systems. That makes IT services firms a route to market for companies such as OpenAI.
• For service providers, generative AI brings risk and upside. Tools such as Codex can reduce some work, while giving firms a way to sell more valuable services.
• Infosys is weaving generative AI into large outsourcing contracts as a productivity tool. The company said it usually supports cost-efficiency deals rather than serving as the only reason a deal happens 1.
• This approach helps Infosys address a split market. It can use coding automation to lower costs for cautious retail clients, while backing more discretionary projects in parts of financial services where spending is picking up 1.