Highlights
- When AI began to take over the world, we were told that there would be some roles that disappeared and new ones that would emerge.
- There’s been a very clear change to job demand in the Salesforce space. However, the new roles that are emerging are few and far between.
- Traditional roles remain, and the new AI tasks are rolled into those traditional roles alongside Salesforce tasks.
Artificial intelligence has dominated the conversation for the last three or four years now. In the Salesforce ecosystem especially, the biggest talking point has always been jobs – would AI eventually replace admins, developers, consultants, and architects? It was, and still is, one of the biggest questions hanging over the industry at the moment.
But if you rewind to 2024 – right in the middle of the AI boom – Salesforce was telling a much more optimistic story. In an article titled “AI Isn’t Taking Your Job – It’s Setting You Up for a Better One”, the cloud giant argued that while some roles would inevitably change, entirely new careers would emerge too. Prompt engineers, AI trainers, AI auditors, AI ethicists, conversation designers, and so on. It was a clear message that stated AI wasn’t just going to reshape work, but create exciting new opportunities for modern workforces as well.
Fast forward to 2026, and AI has undoubtedly changed the way Salesforce professionals work. Yet many of those predicted AI job titles are surprisingly difficult to find. Instead, what seems to have happened is that AI has been folded into jobs we already know, expanding the responsibilities of Salesforce’s most notable roles rather than creating a wave of entirely new careers. So what changed? Were the predictions too optimistic, or has AI transformed Salesforce careers in a way no one forecasted?
The Promise of New AI Jobs
Looking back, it certainly wasn’t just Salesforce predicting new jobs would emerge from AI. Microsoft was talking about an entirely new AI economy at the time, while the World Economic Forum stated that 170 million jobs would be created by 2030, even as 92 million were displaced. There was genuine belief across the industry that AI wouldn’t simply eliminate work, but create new opportunities alongside it.
And there is evidence that happened, at least to an extent. The Wall Street Journal reported that AI contributed to around 640,000 new jobs in the US between 2023 and 2025, with AI-related vacancies steadily increasing.
The catch, however, was that the vast majority of those openings came from a tiny proportion of employers, suggesting the boom wasn’t as widespread as many expected.
Perhaps the best example of this is the prompt engineer role. It quickly became a bit of a poster child for the AI job market, with companies advertising roles focused on designing and refining prompts for large language models.
On paper, it sounded like the perfect example of a brand new profession. In reality, many of these roles now require years of AI or machine learning experience and a deep understanding of model behavior – not simply someone who can write out good prompts.
There’s also evidence that AI models themselves are becoming some of the best prompt engineers out there. Rather than relying on a person to painstakingly find the right wording, researchers have developed systems in which an LLM generates instructions, tests performance, and refines or replaces them.
One study that was boldly titled Large Language Models Are Human-Level Prompt Engineers found that automatically generated prompts matched or outperformed human-written instructions on 19 of 24 language tasks. Google DeepMind’s OPRO technique went further, using an LLM to repeatedly optimize prompts based on previous results – its best prompts beat human-designed alternatives by up to 8% on one maths benchmark and 50% on a collection of difficult reasoning tasks.
Moreover, a 2025 academic study analyzing more than 20,000 LinkedIn job listings found just 72 prompt engineer vacancies. Rather than becoming a mainstream career path, prompt engineering was already being folded into other technical roles. Looking back now, it’s hard not to wonder whether the same thing happened to many of the other AI careers we were promised – only much faster than anyone expected.
Then vs. Now
The reality is that we’ll only see new roles evolve if there is a genuine need for them. There’s no denying the impact of AI tooling on careers, Salesforce and beyond, but some of the changes haven’t played out the way we thought they would.
As the tools have evolved, so have the skill requirements. As skill requirements have evolved, so too has the general AI knowledge that professionals have gained. The more we’re exposed to AI, the more we tend to have a good general knowledge of it.
This is where the majority of the gap has come from – it’s not that the need for AI skills, knowledge, and experience isn’t what we expected it to be. It’s that the AI wisdom that we’ve all gained as the tools have evolved has more than made up for the lack of understanding that we had when they first hit the market.
The solution has become a bit of a problem. We knew there would be a need, and thought it would be as simple as carving out new roles to replace those that were going away. We believed that there would be less of a requirement for traditional Salesforce professional roles and more of a need for AI roles to control and contain what was coming. The reality, as you’ll soon read, didn’t follow this same path.
More of the Same
We incorrectly believed that the Salesforce-specific roles would disappear, and that AI would perform a majority of the work that we used to. We believed, and were sold the idea, that the requirement would be for AI-specific professionals to be deployed over the top of AI that understood and took action on Salesforce.
The reality, at least for now, is slightly different. All the years of Salesforce knowledge that we’ve acquired and developed over the last almost three decades of the CRM giant existing are still extremely valuable.
Yes, we’re absolutely seeing tooling that can understand and do things in the Salesforce space – Agentforce Vibes can read and work with your org’s metadata to build new customizations in record time. Setup with Agentforce allows Salesforce Admins to communicate with their org in natural language, and Agentforce will make changes based on these conversations. We even have third-party solutions like MeshMesh that take all of this to a whole new level with agentic building, documentation, and project organization.
All this being said, Salesforce professionals (admins, developers, architects, and other traditional Salesforce roles) are still valuable, necessary, and critical for Salesforce success. This is despite the warning (and realities) of job loss, and the promise of new AI roles that we have heard about over the years.
The new roles we were promised – like that of a prompt engineer – are simply becoming new duties that are listed on traditional Salesforce roles. Where an admin previously would’ve been expected to know how to create and manage users and permissions, now they’re expected to know how to work with Setup with Agentforce.
Developers previously would’ve been expected to work with and build new Apex Triggers and Classes, but now they’re expected to know how to work with coding agents like Agentforce Vibes, Codex, and Claude Code as well. The roles we thought were coming have become new duties that we’re expected to take in stride.
This may sound harsh, but the reality is that competition for our roles is stronger than ever, and our competition is being met with the same expectations.
The Salesforce job market is a dog-eat-dog world at the moment, and your competition is polishing their AI skills to be more attractive to potential employers.
Were We Wrong?
Calling the predictions around AI-related job roles “wrong” probably isn’t a fair assessment. Most of the major forecasts around AI and employment stretch out to 2030, and we’re only partway through that journey. There’s still every chance that entirely new professions emerge over the next few years as AI becomes more deeply embedded across businesses.
But if we’re judging the predictions based on where we’re at today, the reality does look a lot different than what was expected. AI has absolutely created new work, but much of it hasn’t arrived in the form of brand new job titles.
As mentioned, prompt engineering has become a skill rather than a career; building AI agents is increasingly part of consulting and development work, and AI governance naturally sits alongside architecture and security. Across the board, AI knowledge is quickly becoming an expectation for any Salesforce professional in the ecosystem.
In many ways, it may have been assumed that AI would follow the same pattern as previous technology shifts. CRM gave us Salesforce Admins, Cloud computing created specialist architects, and Cybersecurity became its own discipline with dedicated engineers. AI, at least so far, hasn’t followed that blueprint.
Instead, it seems to be expanding the roles that already existed rather than replacing them with entirely new professions. That may still change by 2030, but for now, the biggest impact of AI on Salesforce careers hasn’t been the jobs it has created, but how drastically it has reshaped the ones we already had.
Where Now?
Even though 2026 doesn’t look the way we thought it would a few years ago, it’s not a sign to let your guard down. Business as usual is well and truly over, and the AI race is well underway.
As a Salesforce professional, you should continue doing what you’ve always done in terms of learning how to properly work with the Salesforce platform and tooling. Your years of experience working with the cloud giant aren’t going to become useless to you – in fact, quite the opposite. While AI tooling is getting more and more capable by the day, there will always be a need for those who understand Salesforce deeply.
Any old AI tool can help you to build an Apex class, but only a skilled professional can ask the right questions to the right people in the business to know why and how it should be built.
However, you should also not neglect your AI upskilling. While Salesforce tooling is important to stay on top of, so too is the emerging suite of AI tools that are becoming more and more prevalent in the tech space (and beyond). There’s no denying that Salesforce is also doubling down on AI in their offerings. They’ve recently rebranded quite a few of their credentials as well to include the Agentforce branding, showing they’re not shying away from their new position in the AI-driven future.
Ultimately, there’s a lot more to keep up with. Salesforce, AI, and, perhaps most importantly, your soft skills will all play a role in the success you find in your career. With AI tooling giving everyone an edge to do more than ever before, you’ll find that demand for your role will likely plummet if it hasn’t already. There will be more professionals fighting for the same job openings. It’s an employer’s market like the Salesforce space has never seen before.
Summary
We’re well and truly into the AI era and are already experiencing its impact on Salesforce careers. There’s no denying that AI is becoming more and more important when working with Salesforce, as with many other technologies. That being said, the new job opportunities believed to have been carved out simply do not exist. At least, in the way we thought they would.
Traditional Salesforce roles will continue to be important and sought after for the foreseeable future. Where the prediction differs from the current reality is that the new jobs that are required for the AI era are being folded into existing roles, rather than creating whole new ones.
Have you noticed a change in the work that you’re required to do in your current role, or did you notice a shift when you took up a new role? Are you finding it difficult to keep on top of the new AI tasks as well as traditional Salesforce tasks? Let us know in the comments.
