Zoom is entering into a strategic partnership with Anthropic that calls for the latter company’s generative AI model to be integrated into Zoom’s Contact Center portfolio.
In an announcement on Tuesday, Zoom also stated that the company had made an investment in Anthropic as part of the collaboration, but did not disclose the amount or terms. Anthropic also has backing from Google’s parent company, Alphabet, and its flagship AI, dubbed Claude, has been described as a chief rival to ChatGPT.
Claude is what Anthropic calls a “constitutional AI” model, meaning that it uses its own AI to track values like “nontoxicity” and “helpfulness” in an effort to avoid AI hallucination — the tendency for generative AI to sometimes make things up — and the use of AI to perform harmful acts. A two-step process provides for, first, an iterative self-critique and tamping down of potentially harmful responses to queries, followed by AI evaluation designed to train a preference model so that Claude can identify better and less harmful responses.
The idea, according to Zoom’s announcement, is to let AI assist call center workers in understanding a given caller’s intent. The system should be able to surface helpful information quickly, and provide insights that can be used to help train call center workers. A planned feature will also use AI to offer those workers resources based on caller requests.
Zoom chief product officer Smita Hashim said that the AI is a key part of the company’s strategy for creating more helpful interactions between call center staff and customers.
“With Claude guiding agents toward trustworthy resolutions and powering self-service for end-users, companies will be able to take customer relationships to another level,” she said in the company announcement. “Partnering with Anthropic also furthers our commitment to providing customers with our federated approach to AI, optimized to deliver outstanding customer experience outcomes.”
Zoom’s in-house AI models are another part of that federated approach, which the company said offers users different options for different applications. Additionally, the new Contact Center AI functionality can incorporate AI models from customers themselves, offering an additional layer of flexibility.
Anthropic has also made inroads with other AI partners, most notably DuckDuckGo, a privacy-focused search engine. DuckDuckGo announced in March that it would integrate Claude into its DuckAssist feature — which also utilizes OpenAI’s Davinci model.
Copyright © 2023 IDG Communications, Inc.
This story originally appeared on Computerworld