Einstein GPT & Salesforce AI Day’s Top Takeaways

Since ChatGPT made its way to the mainstream, businesses of every shape and size have been trying to figure out how to use Generative AI technology to enhance their business.

Enter: Salesforce’s Einstein GPT, the first AI for CRM. Marc Benioff, CEO of Salesforce, took the stage in New York City for Salesforce AI Day, a highly anticipated session. 

Sometimes you just need to “be in the room where it happens'' and NeuraFlash was proud to be one of the few partners who got to experience everything live!

Here’s the TL;DR:

The AI boom (if we can refer to it as such) is in full swing, and the world's #1 CRM company, Salesforce, recognizes the importance of advancing their technology in tandem with the widespread availability of open-source access to these cutting-edge Large Language Models (LLMs).

"I think Artificial Intelligence and Generative AI are probably the most important technologies of our lifetime." - Mark Benioff, Salesforce CEO

Let's breakdown our team's 5 biggest takeaways from this event!

‍Privacy, Data, and Ethics with Einstein GPT 

The issue at hand is of significant concern for all companies, regardless of their size or security measures. Even large, well-established enterprises cannot afford to expose their customer data without careful consideration. However, when it comes to effectively utilizing AI in a CRM, tapping into existing customer data becomes crucial for achieving desired outcomes. While the ethical implications of Einstein GPT require ongoing exploration, Salesforce has previously faced similar challenges and places a high priority on the security of Personally Identifiable Information (PII). To address this, they have introduced the "Trust Layer," a comprehensive approach that specifically addresses data security concerns associated with LLMs. This trust layer serves as a safeguard for sensitive customer and business data, ensuring both its protection and effective usability.

‍Generative AI is a catalyst for positive business outcomes

Companies ranging from AAA to Rossoginal are currently piloting this groundbreaking technology. They have already started witnessing the significant operational benefits, and their use cases have global applicability. These are just a few of the Generative AI use cases. So what are some of the business outcomes you can expect from Generative AI? 

As an outcomes based company we are always thinking about how we can measure the impact! Here are some of the outcomes you would expect by utilizing Einstein GPT:

1. An increase in First Contact Resolution: By providing suggested responses and likely solutions, Generative AI can super-charge the agent to a faster and more accurate resolution, with minimal back-and-forths.
2. Increased Self-Service Capabilities: Generative AI can be used to help guide customers through self-service processes, making them more successful and potentially increasing the breadth of issues that they can complete through self-service.
3. An increase in Agent Efficiency: With suggested responses, auto-creation of email content, suggested knowledge articles, auto-summarization of conversations, and auto-categorization, agents can handle more conversations more quickly and help more customers.
4. A decrease Average Handle Time: By providing suggested responses and likely solutions, Generative AI can super-charge the agent to a faster and more accurate resolution.

The product is going to continue to evolve  ‍

Generative AI has taken the world by storm, it’s only natural that the speed of innovation is going to be just as fast! Generative AI technology is iterative and will continue to evolve, but we do know some of the the features you can expect in the coming releases of Einstein GPT. Take a look at a few of these Generative AI use cases.

• instant text translation, 
• personalized email generation, 
• customized landing pages, 
• comprehensive work summaries, 
• advanced knowledge article surfacing,
• & so much more!

It's all about ‘Augmentation’

Highlighted often by Mark Benioff, was that the central goal of this product is to empower and enhance people's work through the strategic utilization of Einstein GPT. This will pave the way for this cutting-edge technology in streamlining processes and simplifying the lives of agents and sales representatives.

One of the primary objectives of incorporating Einstein GPT is to revolutionize the way tasks are performed, enabling agents and sales representatives to achieve greater efficiency and effectiveness. By leveraging the power of Einstein GPT, you are able to optimize workflows, reduce manual effort, and increase productivity.

This point was further emphasized by Benoiof as he unveiled the entire suite of GPT Clouds all nestled underneath the company’s new AI Cloud. These GPT Clouds are designed to enable businesses across teams and departments to do more with less, offering a comprehensive and easily navigable knowledge base that equips them with the necessary resources to excel in their roles.

But where's the data? 

LLMs represent a departure from traditional data structures, yet the importance of data and its relevance to businesses remains unchanged. Introducing a new data structure brings the question of data storage to the forefront of conversation, it becomes crucial to consider where all the information is being gathered and stored. Unlike traditional AI and data structures that have designated storage locations, often more than one, LLMs don't require that same storage format. But a unified data source is vital for businesses and thus, the successful implementation of Einstein GPT. This is where Salesforce Data Cloud serves as a cornerstone, playing an essential role in establishing a solid foundation for any type of work

Don't want the data there? No worries! With protections built in by the Salesforce trust layer, things like data masking can cover that data. Want to remove a prompt? Don't want to teach the model? No worries! You can delete that prompt before the LLM even learns it, keeping your private information private.

Turning “on” Einstein GPT

We use ChatGPT at the “touch of a button” …there is very little red tape to feed ChatGPT a transcript and say “give me the summary.” However, in order to leverage this technology in your existing business CRM or channels, there will be key features and platform requirements before you can even begin to leverage Einstein GPT. More than just licensing, you’ll also want to ensure your organization’s readiness for adopting this new technology. From your data, to your people, to your process, all have improvements to make to ensure you’ll get the outcomes you desire. 

Here are a couple things to consider:

1. Are you currently using automation like bots and digital channels?
2. Where is your data housed today? Do you have multiple integrations and APIs?
3. Do you have a change management plan to help upskill your team?

NeuraFlash: A leader in AI-driven business value

As a leader in AI-driven business value, we are inspired by the technology advances and the broad visibility Generative AI is experiencing, and we see immediate and ongoing opportunities to help our customers understand, leverage and extract value from these dramatic advancements. NeuraFlash has always been at the center of helping companies be at the forefront of innovation, and we are already working with customers on their vision and questions about how their business can benefit from the next generation of Generative AI..

Want to learn even more about this exciting day? Check out what Salesforce had to say.