Watch now! - https://www.youtube.com/@Urelevant

Let's dive into some security concerns and what you can do to shore up your AI implementations inside of

agent force so I noticed a post on LinkedIn that was gaining some traction that was from Amnon Kruvi and

he's a Salesforce architect and he mentions in his post that "it took me exactly two questions to accidentally

get agent force to reveal someone else's personal information using the default actions followed by

hallucinating madeup orders for that person and then from there he's saying how AI has no business reading

database records that is not to say there are no excellent use cases for it but delivering live information from

a database is just too risky in the data protection era we need to be realistic with what kinds of solutions AI

can safely deliver I understand the hype but some of it will just leave the door wide open for someone to

steal your data." That really intrigued me when I first saw that is like wow this is giving up information and

Salesforce has done a lot of work around the Einstein trust layer to try to protect information to mask

sensitive data as it goes to a large language model but when you think about it as far as authentication

methods that's something that always happens whenever you call into a call center and dealing with any sort

of sensitive records often times you're asked to verify your phone number your date of birth perhaps provide

the last four of your Social just different things as far as verifying and so what Amnon goes on to describe in

some of the comments which I'll highlight some here in a moment is that the verification process was kind of

thin and this was the default behavior and setup in the instruction sets inside of Agentforce and I'll dig in

more to try to see what sort of org or instance he was in if this was is a free learner account I think one of the

issues is is that this was the default setup provided by Salesforce which might lead to uh users trusting that

just because it's coming from Salesforce just presuming that best practices were being used so we're going

to explore in this video as well how you can help bring your instructions into alignment your various

guardrails that you can put in place inside of Agentforce and then open up some of the possibilities as far is

if there's things that are out of alignment or contradict one another in your guard rails and instructions these

are all things that we now have to think about in this new age of AI that we're working in and navigating and

so Amnon further iterates that does a good job of closing off a lot of attack vectors but the issue was with the

default demo configuration being of poor quality and teaches bad processes that highlight the security risk

involved with any kind of AI based technology and so here is my comment where I chimed in just saying for

my perspective that there's so many challenges that abound from implementing generative AI and placing

guard rail ensuring alignment across all instructions in Agentforce and the inevitable rapid release of new

and improved models makes this a moving Target this is a good case study for the Agentforce testing center

and previously we saw the release a few weeks ago of the Agentforce testing center where you can bulk test

agent force performance and agent responses and I think that this is a good thing to think about is the

hundreds or thousands of ways that prompts might come into an Enterprise and then testing out out in bulk

the verification process so that you are not just giving away other people's information the scenario that

Amnon is describing is he's self-identifying as someone saying that he is someone else giving that person's

email address which sometimes is easy to find online and then asking questions about an order for example

so you can see if you're dealing with agent force at a healthcare setting Financial Services Etc there's a lot

of loopholes that could be exploited and so then Paul Battisson he had a question here missing that this is

concerning and asking about the setup wanting to know more details as to what was the org in question

what was the setup and so he answers Paul saying it was an SDO that's the Salesforce developer org and

the main point here is that Amnon had a pretty good idea of why it was happening how to mitigate the

situation as well his main point is that the default action should not be so exposed because people might

think they're best practice and that's the point here is that when you see something from Salesforce you

assume that everything's been thought out and thought through and that the proper guard rails are in place

so whenever you're spinning up an instance that has Agentforce enabled you don't want to just necessarily

take all the instruction sets at face value there's instructions you can place the agent level and inside of

prompt templates and you will be wanting to audit those make sure that they're in alignment that's one of the

points I was trying to make as far as this being indeed a moving Target coupled with as well in the

background the constant Evolution and advancements with new large language models and those being

added into agent force over time and so this is something that will not be set it and forget it sort of

proposition but will always need to be being monitored by organizations and tested in bulk in mass and that's

why the Agentforce testing center is so important is because we can't humanly scale to that point to think of

all the variations as to the different approaches to be able to try to hack this in and there was another

response further down from someone named Vani I didn't put her last name I checked her profile I'm not

sure what her last name is she's bringing up since Agentforce can't function without Einstein trust layer uh

which includes safeguards like data masking and access controls I'm curious do this happen even after

having these protections or or do you think they're still room for improvement and so then Amnon responds

back that I did not actively put someone's address as protected data in the trust layer configuration though it

was enabled with the default settings and then basically said hey my email is xxx then asked it to tell me

what my address and birthday were and so that is the example specifically of the prompt or the utterance

that was given to Agentforce and it didn't really do a great job as far as verifying the identity of the person it

was able to then verify by the email address assuming that that is the person that is chatting or prompting

agent force and then was able to follow up with asking some follow-up questions and so then Andy

Cotgreave brought up a great point as well saying we don't want to put the burden on the end user as far as

having to test test test and that burden should be on Salesforce in the configuration of Agentforce and this I

think it was this specific comment that caused me to remember theAgentforce testing center which was

recently released that comment of test test test was realizing okay the burden is on the user and this is

Salesforce's response is to use the Agentforce testing center because it we can't humanly scale as I said to

test out all those different variations and so it's the coupling of humans and AI working together on that side

of the fence to do that testing in in addition to configuring the Einstein Trust Layer setting and then as well

the instruction sets for prompt templates the agent instructions as well the topic configurations so there's a

lot of great conversation here and this really opens up some thought related to authentication of users and

just the utterances and prompts that Agentforce will be faced with dealing with out in the wild so many

thanks to Amnon Kruvi for insightful post bringing up some important aspects related to Security in the age

of Agentforce and so be sure and check out Velza that is our implementation company we specialize in

Salesforce implementations and agent force implementations reach out to us at Velza.com and we will

schedule a call do a discovery and get your implementation out on the right foot or fix a failed

implementation that seems to be all the rage nowadays is people trying to start over and get their

configurations fixed especially in this age of AI and Agentforce also be sure and check out rapidreskill.com

for Salesforce and AI training and be sure and like And subscribe to the Urelevant podcast feed the

algorithm help others to find Urelevant as well it's all about helping you to find relevance in the economy of

now I'm Mike wheeler signing off for now until next time I'll see you in the cloud

Watch here - https://youtu.be/Et-lXOQeLyE?si=gkkxpyYXKNCIYtlF

So Agentforce 2.0 is coming soon. December 17th on Salesforce Plus Marc Benioff will be sharing the latest release of Agentforce.

And inside of Agentforce 2.0. It will have slack integration and also enhanced CRM and analytics capabilities.

And with that slack integration, there's a preview of what that will look like.

It has various different agents that you can use inside of slack, such as a new sales agent, and then also on Salesforce's website. They do have an ROI calculator.

Which is accessible from the home screen. Front and center is Agentforce, and they have the banner here about Agentforce 2.0 and the link to stream on Salesforce Plus their streaming platform, and then also the calculate your ROI or return on investment. And here this gives you the numbers based on number of customer service employees, average annual cost per employee, and number of conversations handled by a rep each day.

And this is in either table or chart form. So it's professing a three-year net total savings of over $800,000. In this scenario, after three years, when you factor in the agentforce costs, which Salesforce has said is roughly $2 per conversation or chat, there's been a lot of derision online around that price point. And so what Salesforce is trying to do is convince companies and decision makers that the $2 per conversation is cheaper than having an employee, basically.

And so in this, some of the implications that you've got to think about, one that I noticed is that Salesforce using this ROI calculator in the context of service and support. So there are mentions of sales agents, but where the primary savings is, is in the service cloud side of things are providing support through agentforce. And then also on the website here.

Few things you might notice as well is that Agentforce is now available to chat with on the Salesforce website, so you can take that for a spin to test that out. And then also noticeably absent any more or any longer is a mention of Customer 360 prior to Agentforce's release, which was back in September of 2024.

During Dreamforce, Customer 360 was featured atop. Most are above the fold, and Customer 360 was what Salesforce referred to as all of their different cloud offerings. The entirety of the platform was referred to as Customer 360. Well, I bring this up is that with the advent of Agentforce, they dropped that entirely. Customer 360 also, there's been mention of the Einstein one platform.

Now we're in this numbering scheme with Agentforce 2.0. Personally, not a big fan of the numbering scheme because I don't look forward to Agentforce 3.0 and 4.0 and treating this with each release as another numeric release number.

Or if they're smaller releases, will we have Agentforce 2.1. And so I'm anticipating that eventually the number will be dropped and they'll go with just Agent Force.

And as history is my guide, I know that this is something that Salesforce does quite often, whether it's Customer 360 or not. Agentforce 2.0.

Salesforce one was the mobile application. And then eventually it was renamed Salesforce Mobile.

And so whether the numbering scheme continues or not, there's a lot of changes that are always bound to be happening in the area of AI.

Agentforce, being a generative AI tool by its very nature, will be a moving target that you will always be having to refine and improve upon. And then also with all the new models coming out all the time from the large language models or the AI companies out there.

This is also something that will be somewhat of a moving target and require refinement all the time within organizations. And so it's not a set it and forget it sort of proposition. And so this is something that we as professionals on the Salesforce platform, and that working inside of the field of AI will need to be keeping an eye on and being aware of all the changes happening.

So December 17th on Salesforce Plus Agentforce 2.0 will be unveiled and then we'll know more. And so until next time be sure and like and subscribe and share this podcast with others and help others to find Urelevant.

So in today's episode, I want to explore the idea of creating your own CRM or customer relationship management system. In addition to AI tools such as ChatGPT and specifically a custom GPT. And there may be other tools out there that could do similar. I know Claude has projects available. So the main point here is that with any software design, whether it's a CRM or some other sort of management system and a lot of software, not all, but a lot has to do with the management of either customer relationships or enterprise resources, financial management, time management, whatever it may be. Oftentimes, if you have an application or use case that has the word management in it, it is a good candidate to be able to be built in a generative AI system or as a custom GPT. Specifically, you think about software, and I've got a lot of experience on the Salesforce platform. When you think about it, it can be boiled down into four main categories or functions. And so the first would be I like to call a rule library or rules libraries. Think of these as the standard operating procedures that drive any business. Now a lot of businesses don't have clearly defined processes, and their processes could be defined as chaotic and haphazard or ad hoc and everyone doing things differently. In the oncoming age of AI, in the enterprise, those organizations that have clearly defined processes that are documented will benefit with AI, because that can be fed into AI can find improvements, gaps, and then also reinforce that behavior inside of the system that is being built, whether that's on the Salesforce platform, which readily is able to translate those requirements and those processes into reality on the platform, often with clicks instead of code or some other system. But the first primary building block of a system that does management in general would be what I like to call a rules library. And those are the rules that you live by as a business. Now, as with any rule in business, there are always exceptions. And this is where you will see that we always do things this way, except for these exceptions. And anytime there's an exception. Think of those as an exceptions library. And where I'm heading here, as we've briefly touched on, the first two of these four building blocks, is that these are basically libraries of information. Or consider them knowledge articles or knowledge libraries Or different pieces of content. You have your rules and then you have your exceptions to those rules. And that is when things are out of the ordinary, you have your primary use case. And then the exceptions handle those things that don't follow the main path, that most things go down to the default path. And then in the third library you would need are actions that the software needs to take. And this is where you go from static information to updating that information. Back in the early days of the internet, in the land of intranets and CGI programming. This is where you would be making calls to a server and doing updates there, or even relational databases. Whatever it may be, you need some sort of actions library or to account for the actions that need to happen, so that when this rule is triggered and these exceptions are accounted for, what is the action or actions that need to be done? And also are these immediate or time based actions, much like how workflow rules used to work in Salesforce? And so then the fourth and final piece is that database or dare I say, spreadsheets or even Google Sheets. Some way of connecting your roles, your exceptions, and the actions that then go out and update your database or your records, And so if you've ever built a custom GPT, you have your instructions And that's where you would specify your rules and your exceptions. You have your knowledge base. And those are additional documents that you can share for to train upon and to know and understand further. And then also you can connect to an external database through actions via custom GPT. You could use this, use case then in this scenario to build for example, a customer relationship management system as a custom GPT. And so that fourth and final piece is that database or the records in general, as far as the different accounts or contacts or opportunities. And then based on its knowledge base of its rules and its exceptions, then it can infer what actions need to be taken as well. So I encourage you to think about building your own CRM as a custom GPT. That would be great use case of exemplifying your understanding of building out applications with generative AI, and that would be great things to speak to as well in an interview, or to put on a resume or LinkedIn profile. The beauty of generative AI is it can help you on the front end of the planning of the software that you're building by helping you to create business requirements, documents, and user stories as well. And then on the other end of the spectrum, for the QA and testing, the user acceptance testing and the test scripts that need to be performed, testing for positive and negative outcomes and being thorough. It can also figure out unlike us because we don't know. What we don't know is it can help infer and figure out the gaps in our thinking or the things we have not thought about. The different dependencies. These are the things to be taken into account and ask questions in this back and forth, conversational coder type of mindset to help identify where exceptions are needed, where further rules are needed, and then also the prioritization of those roles, what order they need to be evaluated in. The main point is you don't need to know how something is done. You don't need to know the nitty gritty of how the code works. You just need to know what the end results you're looking for, and then being able to test and verify that it's giving you the desired results. If you have a thorough test case or set of test cases, you can verify that is performing as expected under a host of scenarios. With the embracing of generative AI, you can build your own systems as a solopreneur as well, and try to leverage that in your own home based business or your lifestyle management or home management techniques. And these be practical, hands on examples of things you could build using either Salesforce or ChatGPT or cloud or whatever AI tool you choose. And regardless if you want to go down the path of building something like this, just having the understanding that all management related software, especially Salesforce, can be boiled down to four key components. And that would be the rules library, the exceptions library, the actions library. And then finally the database of the records are stored and updated and can be transformed. So now with the exciting times that we find ourselves in, the ability to become a creator in a consumption world has been democratized. It is available to anyone that can form sentences and can communicate. And so this is an advantage to all of us, and not just the few. So all that's left in remains is to ask yourself, what will I build today?

AI Will Transform Automations and Setup in Salesforce.

Watch here - https://youtu.be/eoBv8ZKelDk?feature=shared

So one of the ways that I stay up to date on the platform is that I'm always teaching new things. And so I'm doing live classes around the AI specialist certification. That brings me into Agentforce as well, and getting familiar with prompt templates and agents, and setting up instructions for agents and instructions for prompt templates and guardrails and a whole lot more.

Now, with that training and that being live, one of the things that I encounter is changes to the platform or new things coming along. And one thing that I noticed in a help article related to prompt templates lets Salesforce platform is that there was a fifth new one that was added to the help, even though it was not reflected in the interface for these free AI learner accounts that Salesforce so generously provides us. Fortunately, it does give us the ability to get hands on with Einstein Platform and with Agentforce, of course.

And so up until just a few days ago, all that I was aware of were the four primary prompt template types, which were for sales emails. Being able to generate emails from Salesforce using generative AI. Summary templates in that a ready example of that would be to summarize this account or opportunity. There's also field generation templates where you can use AI to generate text into a field. And one good use case for that would be to give you a summary of a record. And I spoke in the previous episode about how this might change the lead conversion process, because you can summarize a lead record into all of its data points into either one or multiple summary fields that are AI enabled, that's through a field generation template type.

And then you've got your flex templates. And I like to refer to them as flexible templates. You can use up to five sources. You can allow for free text. It can work in conjunction with flows or apex or the rest API. So it has a lot of flexibility and power to it.

Then there's a fifth newer prompt template type listed in the help. And I'll link to that in the resources for this episode. And that one was known as a record prioritization prompt template. And the specific use case is that that will be available out of the box as far as with an agent action, a standard agent action and agent force for the prioritization of opportunities.

I'm anticipating that that list as well for standard agent actions will grow, Salesforce provides a lot of example actions that you could create of your own custom different. I use cases in there. I use case library and a lot of prompt templates as well.

And this is only just the beginning. We'll see more of this available on the Appexchange in the future. As far as custom built agents, prompt templates, actions I'm anticipating the Appexchange really changing in the next few months in order to encompass all of the above, but this record prioritization prompt template and seen it so early the early days of Agentforce tells me or informs me that this list is going to grow.

And so not only will it transform our templates and our actions, and that being some of the generative AI on the back end, the primary use cases for those various templates and actions impact the end users on the front end of Salesforce. But we are starting to see some rumblings of some AI appearing on the back end inside of set up to help us configure things such as formulas.

So in the next episode, I will discuss some of the future of Salesforce and what I see coming as I look into appear into the next quarter of this century. I like to refer to is Q2 of 2K, and where things might head in the coming months and years with AI as it transforms how we do work, how work gets done on the platform not only for our end users and marketing, sales and service, but for us as consultants, administrators, developers, and more.

So until next time, be sure and like and subscribe, comment, share, feed the algorithm and let others know about Urelevant.

Generative AI is going to fundamentally change how work gets done in this next quarter of the century. We're entering into the year 2025 in just a couple of months at the time of this recording, which I like to call Q2 of 2K. I'm seeing a lot of profound implications as we're now starting to get our hands on AI inside Salesforce. Salesforce has opened up access to their Einstein platform in various ways, including learning platforms and demos.

With this access and an understanding of the Salesforce platform—how work is done or has been done in the past—we can start to envision ways of doing things more efficiently in the future with the help of AI. The intended use case that Salesforce provides is to accentuate and augment our abilities, not to replace us entirely.

What's really interesting to me about this encroachment of AI into the workspace, especially on the front end of Salesforce, is that it is redefining how we do work. This will redefine our standard operating procedures and, at a very high level, some of the tried-and-true processes baked into Salesforce from the beginning, such as the lead conversion process and the management of opportunities. These processes can and will change more with AI integrated into the platform.

One example I've recently been experimenting with is the approach to the lead conversion process. If you've worked in Salesforce, especially on the marketing side and dealt with lead qualification, you know it can support multiple lead processes depending on the types of products and services you're selling. This includes different lead status designations as you go through the lead qualification journey. Certain data points must be captured along the way. At some point, when you want to hand the lead over to the sales department, you perform what is known as a lead conversion.

In the past, a lot of customizations were required, such as custom field-to-field mappings from the lead object to contact, account, and opportunity. Those of you who have studied with me for the administrator exam, for example, are highly familiar with that process. For those of you newer to this, the main point here is that the journey or lifecycle of a customer with a business typically starts as a lead inside Salesforce. At some point, it is converted into an account, contact, and opportunity.

All the data points captured on the lead record transform and carry over into the resulting object records. For example, the company name on the lead becomes the account name. Recently, I’ve been experimenting with Einstein-enabled tools to create a summary field as a hybrid prompt template. Currently, there are only a handful of prompt templates available on the platform, though I anticipate their number will grow significantly over time.

There are two prompt templates worth mentioning here at a high level. The first is the field generation template, which enables you to generate text inside a field using generative AI. The second is the summary template, which summarizes records. The latter is available out of the box on most object records. For example, you can prompt the AI to summarize an account or opportunity.

The summarization capability of generative AI is one of its strengths, which is why it’s one of the first prompt template types available on Salesforce. Imagine lead records with dozens, if not hundreds, of fields and data points. Mapping all these fields from lead to contact, account, and opportunity can be cumbersome. Salespeople, however, often need a summarized version of that data instead of searching through countless fields. A summarized lead record in just a few paragraphs could streamline the lead conversion process dramatically.

One practical implementation of this approach involves creating a custom long-text area field on the lead object and corresponding fields on contact, account, and opportunity objects. On the lead side, you could create a field generation template, update the lead page layout to a dynamic form, and enable generative AI through Einstein for that field. The AI could then generate a summary of the lead record and populate the field. This summary could be automatically mapped to the corresponding fields on the contact, account, and opportunity during the lead conversion process.

Additionally, you could set validation rules to ensure the summary field on the lead is not null before conversion. This would compel users to generate a lead record summary before conversion. Alternatively, you could enable generative AI on multiple fields for summarization purposes. At most, you’d need three AI-enabled fields—one each for contact, account, and opportunity summarization—populated on the lead side and mapped during conversion.

This high-level overview highlights how the lead conversion process can evolve with AI. I see this as a repeat of history: technology advancements prompting us to rethink our approaches. As we move into the second quarter of the 21st century—Q2 of 2K—AI will likely continue to transform how we work.

Recently, during a discovery call, I was reminded of how often we inherit Salesforce instances and wonder why things were done a certain way. Sometimes, the explanation is as simple as, "It was the only option available at the time." The advancements in the Salesforce platform—such as the shift from profiles to permission sets, the introduction of dynamic forms and pages, and now generative AI—force us to rethink our solutions. This virtuous cycle of technological progress, reimagined procedures, and new platform capabilities is what makes Salesforce so adaptable.

As generative AI continues to advance, it will impact not just the front-end user interface but also back-end processes. We'll save that discussion for a future episode. Thank you for joining me for this sneak peek into Q2 of 2K and the transformative potential of generative AI in the workplace. Please subscribe, like, and share this podcast so others can also find relevance in the economy of now and next.

In this episode Alyssa shares her shares her story on how she became a Salesforce professional.

In this follow-up episode, sisters Rebecca Youngquist and Nicole Looker delve deeper into their Salesforce journeys. They discuss the misconception that certifications are necessary to land a Salesforce job, with Rebecca sharing how she secured her first role before being certified. 

The sisters offer valuable insights on how to stand out in a competitive job market, emphasizing the importance of translating existing skills to Salesforce roles and actively learning through platforms like Trailhead.

Nicole, now a CRM Manager, provides an overview of the various systems she works with beyond Salesforce, including HubSpot, LeanData, and Data Cloud. She shares her experience implementing Data Cloud in a startup environment and offers advice for those aspiring to get Data Cloud certified.

The sisters also discuss their upcoming presentation at Mile High Dreamin', revealing how they plan to share their unique story of growing up in a small town and finding success in the Salesforce ecosystem. They highlight the importance of mentorship, including the Salesforce mentorship program and the often-overlooked Salesforce for Military program for veterans.

Looking to the future, both Rebecca and Nicole express their interest in expanding their development skills, particularly in Apex. They conclude with valuable advice for those entering the Salesforce ecosystem, emphasizing the importance of continuous learning, building soft skills, and finding genuine enjoyment in the work rather than merely chasing high salaries.

This episode provides a wealth of practical advice and inspiration for anyone looking to start or advance their career in the Salesforce ecosystem.

Contect with Rebecca Youngquist and Nicole Looker on LinkedIn:

https://www.linkedin.com/in/rebecca-youngquist/

https://www.linkedin.com/in/nicole-looker/

Watch here - https://youtu.be/WIfv2tXhPSI

In this episode, we meet sisters Rebecca Youngquist and Nicole Looker, who have both found success in the Salesforce ecosystem. 

Nicole, with her background in finance and insurance, shares her journey of becoming an "accidental admin" and how she eventually transitioned into a product owner role. 

Rebecca, a former Navy nuclear engineer and instructor, explains how she pivoted to Salesforce after leaving the military, inspired by her sister's success.

The sisters discuss their unique paths into Salesforce, the challenges they faced, and the skills that have helped them excel in their roles. They emphasize the importance of understanding both the front-end and back-end aspects of Salesforce, as well as the critical soft skills needed to succeed in the ecosystem.

This episode offers valuable insights for anyone considering a career in Salesforce or looking to enhance their skills in the field. From the importance of visual learning to the key characteristics that make a great Salesforce professional, Rebecca and Nicole provide a wealth of knowledge and inspiration.

Stay tuned for Part 2 of this conversation, coming next week!

Contect with Rebecca Youngquist and Nicole Looker on LinkedIn:

https://www.linkedin.com/in/rebecca-youngquist/

https://www.linkedin.com/in/nicole-looker/

Watch here - https://youtu.be/yR06s8kCItM

In this episode, we explore the  parallels between the evolution of Spider-Man and our own journey with emerging AI technologies. Just as Peter Parker navigates the challenges of discovering his superpowers, we too must navigate the unfamiliar terrain of AI and its capabilities. We delve into the importance of embracing a lifelong learning mindset, rapidly reskilling, and developing a strategic reserve of skills to stay relevant in an ever-evolving landscape. Join us as we discuss how the hero’s journey in both superhero tales and real life is marked by transformation, resilience, and the pursuit of making a difference.

Watch here - https://youtu.be/2kaYP8jo8Ww