With new applications for generative AI like Slack GPT and Einstein GPT, Salesforce is extending AI automation capabilities like ChatGPT across its CRM platform.
At the Salesforce World Tour in New York on Thursday, the customer relationship management giant unveiled broad plans to inject generative, large-scale language model artificial intelligence capabilities into platforms like Einstein, Slack, and Sales Cloud. .
This technology enables AI-powered emails, qualified leads, and sales plans. Sarah Franklin, Chief Marketing Officer at Salesforce, took the stage at her Javits Center in New York to explain the risks and benefits of AI in CRM, explaining that despite guardrails and “humans in the loop,” sales and her said to advance generative AI for CRM. protocol.
Jump to:
Slack GPT is third LLM in communications platform
Salesforce announced it is bringing natural language model AI to Slack. This enables low-code AI capabilities, different language models, and the ability to add customer data insights from Customer 360 and Data Cloud.
Rob Seaman, who heads Slack’s enterprise product team, said Slack now has three LLM capabilities.
- Slack GPT consists of native capabilities that leverage LLM within Slack to summarize, compose messages, compose texts, and more.
- Integration of Salesforce LLM Einstein GPT into Slack.
- Based on ChatGPT, it is an AI-enabled platform designed with OpenAI that allows you to build workflows that call any LLM of your choice.
The latter option is similar to OpenAI or Anthropic (Figure A).
Figure A
Seaman said AI integration will enable end users and third parties to build workflows, apps and microservices around features like summarization and transcription.
Last week, Salesforce made its AI-enabled platform designed with OpenAI available to developers of related apps and services. Seaman says Einstein GPT is available in beta, and the company is currently prototyping his Einstein GPT integration within Slack (Figure B).
Figure B
Usability and security are the goals
According to Seaman, the end-user experience and security are the main focuses of the development and deployment process that injects LLM into Slack.
“A lot of what we want to do in Slack is one of our product principles, don’t make the user think. So when these things are natively built into Slack, we want it to be a distinct experience. says Seaman. “Also, many of the most valuable information companies have in the form of knowledge and communication is in Slack.”
Seaman said the company is working on ways to leverage models of that data without the data actually moving and being stored in third-party data sources (Figure C). “Our corporate customers wouldn’t accept that.”
Figure C
look: At the last NYC World Tour Stop, Salesforce spotlighted Customer 360 with its integration of Slack, Tableau, and Customer 360 (TechRepublic).
As part of this process, Salesforce has created data storage and no training contracts with LLM entities such as OpenAI. Mixed with other data sources for training purposes.
As previously reported by TechRepublic, the company also released guidelines earlier this year to reduce bias in AI models regarding verifiability, safety, honesty, empowerment, and sustainability.
Diversified LLM strategy and large app ecosystem
According to Seaman, there have been about 4,000 third-party apps built on the Slack platform in the last two months, specifically working with LLM, including many customer-built apps.
“There are said to be 2,600 direct apps, but there are far more apps than users have created themselves,” Seaman said.
The ChatGPT and Claude apps are example use cases for summaries and drafts, enabling questions and natural language prompts. Can I get a prospecting email template to send to some tech customers?”
Einstein GPT
In March of this year, Salesforce broke the ChatGPT-generated AI conundrum for the first time by announcing the addition of ChatGPT capabilities to its Einstein CRM AI platform. Among other things, it includes:
- Ability to auto-generate prospect emails based on CRM data.
- Automate lead conversion and pre-screen web visitors with a tool called Einstein Bots for Sales.
- Multilingual transcription and conversational AI analytics.
Employees want to use AI to reduce unproductive tasks
According to Slack’s State of Work report, 84% of senior IT leaders say generative AI will help their organizations better serve their customers, and 67% of those leaders expect to made generative AI a priority for their business.
The report found that desk workers whose companies employ AI are 90% more likely to report higher productivity than those who don’t. Those who use automation at work report saving an average of 3.6 hours per week.
During the world tour, Salesforce also announced plans to work with Accenture to accelerate the deployment of generative AI for CRM. The companies will jointly establish a generative AI acceleration hub to provide organizations with the technology and experiences they need to scale Einstein GPT to improve employee productivity and transform customer experience, according to Salesforce. intend to do something.
