# Mphasis 於 5 月 27 日推出企業代理平台 Mphasis Tria，協助客戶從人工智慧實驗轉向成果導向的決策應用。

*genai, business · news · 2026-05-27 · Moneycontrol*

## Key points

- Mphasis 推出專注於代理式 AI 成果的企業代理平台 Mphasis Tria。
- 公司在兩年內投入近 1.5% 營收開發 Tria。
- 企業代理目標為能自主決策且無需人為介入或產生幻覺的 AI 代理人。
- Mphasis 預期成果導向合約將大幅成長，大型合約平均規模提升至 7400 萬美元。
- 兩條新產品線 Mphasis Modernize 與 Mphasis Optimize 將在轉型專案中部署 Tria 平台。

Mid-tier IT services firm Mphasis on May 27 launched Mphasis Tria, an enterprise agency platform that will help customers to move from artificial intelligence (AI) experimentation to outcome-based decision making use cases of agentic AI, said CEO and MD Nitin Rakesh. The company has invested nearly 1.5 percent of its revenue in the past two years to build the platform that will be central to its AI strategy over the next two to three years. According to Rakesh, Mphasis is creating a new category of “enterprise agency,” adding that he has no interest in agentic AI. The company also introduced two new product lines — Mphasis Modernize and Mphasis Optimize — through which the platform will be deployed across enterprise transformation projects. What is enterprise agency? Speaking to the media during the launch of the platform, Rakesh described enterprise agency as a mechanism to train and build AI agents that could autonomously make decisions without human intervention or possibility of hallucinations that is found in current enterprise-grade AI products. “To drive agency, the AI needs to have the ability to make a decision autonomously, which is seeing gap at present…Agentic AI is a tool stack, enterprise agency is more about what can I do better through AI,” he said. He added that while every company has access to LLMs, have invested in innovation labs and have an AI strategy, “they don’t have is the ability to drive ‘agency’ at an enterprise level.” Rakesh described agentic AI as a “hammer looking for a nail.” Mphasis business roadmap Going forward, this will largely drive the IT firm’s revenue mix as it shifts towards outcome-based offerings compared to the present human input based execution. “This will change the shape of our revenue over time. We are calling at FY27 as a foundation of year as we roll out the setup. We should be able to share metrics from FY28. We will also have to reorganise ourselves to some extent because we will be elevating the product and platform work and have forward deployed engineering and solutions of that sitting between the product and the marketing,” said Rakesh. He added, “We are not just playing the game of pure IT services, or we are not trying to become a software company. We are still very much committed to driving outcomes for complex business problems. The way we delivered them is evolving.” At present the contribution from outcome-based deals remain small in the overall revenue mix, but Rakesh expects this to grow eventually. As of FY26, the company has sold 68% more deals than FY25. The average large deals sizes grew from $52 million to $74 million, on the basis of the AI stack it has built and the new IPs launched.

**Companies:** Mphasis

[Read the full story on Moneycontrol](https://www.moneycontrol.com/news/business/information-technology/mphasis-bets-on-outcome-based-deals-invests-1-5-of-revenue-into-enterprise-ai-agency-platform-13933240.html)

---

Canonical: https://newsio.io/zh-TW/n/4888b99c-4fd8-4032-906a-b5904b98b8c2/mphasis-5-27-mphasis-tria-1-5
Summarized by Newsio from Moneycontrol. https://newsio.io/how-it-works