If you’ve been hiding under a rock in the last months, then you might have missed the continuing job loss announcements on news feeds. LinkedIn, Amazon, Meta, Microsoft. The announcements keep coming, and the real lives of real employees are being badly affected. I think it’s time we took stock of where we are and the reality of how we need to pivot to protect our careers.
So what does all of this mean, and how should we be preparing ourselves?
Let us start with the macro picture. Central banks do not often comment on technology trends, but when they do, you should listen. Jerome Powell, Chair of the US Federal Reserve, has acknowledged they are watching AI-related labour effects closely, including scenarios where productivity rises but net new jobs lag. That is central-bank speak for: this could change hiring, wages, and retraining at scale.
On the ground, recruiters are already naming what they see. One I follow calls it “the great restructuring.” Translation. Titles are not disappearing uniformly. Tasks are being re-bundled. Teams are being rebuilt around systems thinking, data literacy, and human oversight of AI.
Now to big tech leadership. There is a lot of performative optimism about AI, but some leaders are unusually frank. Even Microsoft founder Bill Gates says we need to be honest about the impact of AI on jobs. “Honesty matters because sugar-coating delays personal planning.”
And this is not just Western tech. Financial services leaders in Asia are saying the quiet part out loud. The CEO of Southeast Asia’s largest bank has publicly said her own role is at risk from AI. Whether or not that is literally true, it signals a cultural shift in how leaders frame automation and accountability.
So what do we do with all this. Let us be clear-eyed.
Three truths and three traps
Three truths.
- AI is already removing tasks, not only roles. If 30 percent of your workload is template writing, routine analysis or status reporting, expect it to compress. Managers will ask why it still takes a day.
- Two. Organisations will keep hiring, but for different blends of skills. Think domain expertise plus data handling, prompt design, QA of model outputs, governance and the ability to ship workflows, not just create content.
- Policy is lagging. The EU AI Act will shape risk processes, but it will not rescue obsolete workflows. Your best defence is capability, not lobbying your job title into permanence.
Three traps to avoid
- Waiting for your employer to reskill you at the pace you need. Learning budgets exist, but the window is now.
- Chasing tool-of-the-week hype. Master one stack end-to-end for your job. Depth beats dabbling.
- Confusing busyness with value. If your calendar is full of tasks a model can do, your risk is high.
Here is your six-step pivot plan
Think of it as a one-week sprint.
Step 1. Task audit. List your top twenty recurring tasks. Label each as automated, accelerate,d or human-only. If you cannot justify “human-only” in one sentence, it is not.
Step 2. Evidence pack. For three high-volume tasks, run an A/B test. Your current method versus an AI-assisted workflow. Measure quality, speed, and defects. Keep screenshots and timings.
Step 3. Ship a workflow. Build a simple, auditable process around one task using a tool your IT team allows. Include prompts, inputs, outputs, a review checklis,t and a handoff. Make it easy for someone else to run.
Step 4. Add a control. Document risk points and your mitigation. Bias checks, data privacy, brand voice guardrails, and a human sign-off. This is how you become the safe pair of hands.
Step 5. Reframe your role. Rewrite your job in outcomes, not activities. “I reduce campaign cycle time by 40 percent while improving conversion” is harder to cut than “I write social posts.”
Step 6. Publish internally. Share your workflow, results, and controls with your manager and one adjacent team. Offer to teach. People who codify and transfer capability become keystones in the new structure.
If you lead a team, here is a quick operating model to keep people and ethics at the centre:
- Weekly AI stand-up. What did we automate, what did we break, what needs human review?
- A shared prompt library with version control and quality notes.
- Red-team sessions where someone tries to make the workflow fail.
- A clear rule on disclosure to customers or citizens when AI is used to generate content or decisions.
Let me address a common fear. “What if I make myself redundant by automating my tasks?” You are not automating yourself out of work. You are automating yourself into a different level of work. People who lead the change tend to stay to manage it. People who resist the change invite it to happen to them.
Final thought
The story is not “AI kills jobs” or “AI creates jobs.” The story is transitions. Transitions are messy, uneven, and real. Your goal is to shorten your personal transition time. Audit your tasks. Ship one workflow. Show the evidence. Repeat.
If today’s episode helped, share it with a colleague who is quietly worried but ready to act. You can find links to the clips in the show notes. I am JoJo, and this is AI SIX. See you next time.
About the Podcast Host
Joanne Sweeney is a specialist in public-sector marketing and digital communications. She’s the author of Public Sector Marketing Pro and the creator of the Professional Diploma in AI for Public Sector Marketing. With years of experience in training, consulting, and advocating for digital transformation, Joanne provides insight and guidance for government agencies navigating the digital landscape. She is also the host of the award-winning Public Sector Marketing Show podcast.
