Salesforce Offers ‘Open’ Generative AI Suite

Salesforce on Monday announced an suite of generative AI tools called AI Cloud aimed at bringing an “open” approach to the intensely competitive market.

The suite, planned to launch later this year, includes Salesforce’s own generative AI tools, integrated into Salesforce’s flagship products such as Data Cloud, Tableau, Flow and MuleSoft, powering such features as code generation, marketing email creation and business process automation.

It also includes AI offerings from partners including Amazon Web Services, Anthropic, Cohere and OpenAI, allowing users to make use of whichever model best suits a particular use case, Salesforce said.

Customers can bring custom-trained models to the platform while storing data on their own platform, the company said.

Open platform

The announcement furthers Salesforce’s strategy, which it first discussed in March, of introducing generative AI tools such as ChatGPT across the company’s product line.

Salesforce’s own AI models include Sales GPT, Service GPT, Marketing GPT, Commerce GPT, Slack GPT, Tableau GPT, Flow GPT and Apex GPT.

Some are available now, including Slack GPT, Commerce GPT, Sales GPT and Service GPT, with the rest set to launch as early as this month, except for Flow GPT, which is planned for October.

Sales GPT, for instance, can create personalised emails while Marketing GPT and Commerce GPT can generate audience segments for targeting and can tailor product descriptions to each buyer based on their customer data.

Data redaction

The company said it is working to incude an image-generation tool similar to OpenAI’s Dall-E but first needs to address issues such as copyright.

Cloud AI is to include Einstein Trust Layer, an AI moderation and redaction service that scans prompts for sensitive data, such as customer purchase orders and phone numbers, and removes the information before it reaches the model.

The tool is designed to safeguard corporate data at a time when companies including Amazon, Goldman Sachs, and Verizon have banned or restricted the use of generative AI out of concerns the models may be hoovering up valuable information embedded in users’ prompts.