Inflection AI, the company behind the popular Pi chatbot, has launched Inflection for Enterprise, a new AI system tailored to meet the needs of large enterprises. Inflection for Enterprise is built around Inflection 3.0 model family and a robust ecosystem designed to empower companies to fully own their AI capabilities.
Founded in 2022 by Mustafa Suleyman, Karen Simonyan, and Reid Hoffman, Inflection AI quickly established itself as a leader in empathetic and user-centric AI technologies. However, earlier this year, the company underwent significant changes as Suleyman and Simonyan departed to lead Microsoft AI, prompting Inflection to refocus on its new “AI studio” initiative. Today marks the first substantial output from that pivot, as Inflection unveils an enterprise-grade system designed for business-specific customization and robust security.
During a hands-on demonstration, Inflection representatives highlighted the system’s ability to seamlessly integrate with existing enterprise workflows while maintaining strict data security protocols. On the backend, the company is leveraging a strategic partnership with UiPath to provide and necessary automation services.
COO Ted Shelton spoke passionately about Inflection’s offering, emphasizing the critical need for companies to own their AI. “Companies need much more than just a model,” Shelton explained. “Inflection is providing the necessary tools and infrastructure for businesses to create solutions that are customized to a their needs, grounded in their data, and aligned with their internal policies.”
Inflection for Enterprise aims to solve the common issues companies face when integrating AI—from generic, ill-fitted chatbots to concerns over data security. Sean White, who took over as CEO in March, said the company has worked closely with CTOs and CEOs to address these pain points.
Unlike off-the-shelf models, Inflection’s solution is fully ownable, allowing enterprises to fine-tune AI to their unique business needs while maintaining full control over their data and infrastructure. This can be implemented on-premises, in the cloud, or via a hybrid setup, enabling flexibility without compromising security.
The system also incorporates Intel’s Gaudi 3 AI accelerators and Intel Tiber AI Cloud, promising twice the price performance compared to similar offerings. Additionally, Inflection’s collaboration with Intel positions the company as a significant player in the evolving AI ecosystem, helping enterprises achieve faster AI deployments at reduced costs.
Beyond offering a customizable model, Inflection for Enterprise introduces a proprietary fine-tuning system, focusing on reinforcement learning from employee feedback. This ensures that the AI aligns with each company’s ethos, culture, and operational requirements. The goal, according to COO Ted Shelton, is to shift from basic chatbot functionalities to more “agentic” capabilities that deeply embed into enterprise workflows, driving both organizational and individual performance.
Inflection 3.0, the core of the enterprise offering, comprises two distinct models: Pi (3.0) and Productivity (3.0). The Pi model powers the overall Pi experience, focusing on backstory, emotional intelligence, productivity, and safety. It excels in scenarios such as customer support chatbots.
The Productivity model, on the other hand, is optimized for following instructions and is better suited for tasks requiring JSON output or precise adherence to provided guidelines. Both models feature an 8k context window.
Inflection says the AI system will be able learn over time, meaning users can experiment with various prompts and conversation styles. The Pi model, in particular, will mirror the user’s tone and style, including the use of emojis, making interactions more personalized and adaptive.
Inflection also launched a commercial API alongside Inflection for Enterprise, enabling developers to build advanced conversational AI applications that integrate into enterprise environments.
The pricing structure for the API is inline with competitors like Cohere, with both Pi and Productivity models priced at $2.50 per 1M input tokens and $10 per 1M output tokens.
Today’s launch is an impressive pivot for the two year old startup. With its advanced models, flexible deployment options, and focus on empowering businesses to own their AI capabilities, Inflection for Enterprise will immediately be a top option for businesses to consider.
