Senior leaders across the creative and media industries are walking a tightrope. AI cuts cost today. The fastest cut on the table is the junior roles, and that same cut removes the rung every future senior climbs. Entry-level hiring across tech has fallen around 50% since 2019 (SignalFire, 2025). With AI now handling most graduate-level work, the fall is speeding up. The industries are expanding output while quietly closing the door that builds their own leaders.
Key Takeaways
- Entry-level tech hiring has dropped roughly 50% since 2019, and AI handling graduate-level work is speeding the decline (SignalFire)
- VFX studios feel it hardest, where AI now does the rotoscoping and basic compositing that once taught the craft
- Gaming has the highest AI integration in the sector, and the average age of technical hires has been climbing since 2021
- The companies cutting junior roles to protect this quarter are building a senior talent shortage for the next five to seven years
- The agencies getting it right run apprenticeships alongside AI adoption rather than treating the two as a choice
Where AI hiring hits creative production hardest
VFX studios sit at the sharp end. AI now handles the rotoscoping, object removal and basic compositing that used to be where a junior learned the trade (Roland Berger, 2024). Studios report productivity gains of 20% to 65%. Those gains arrive with the career ladder pulled out from under them.
The maths forces the hand. Visual effects eat 20% to 25% of a production budget. Complex shots run up to $46,000 each, and a filming day costs £8,000 to £38,000 (Filmustage, 2025). Streaming services are holding content-spend growth under 10% (BDO, 2025). So AI stops being optional. Productions lean on AI pre-visualisation to cut post timelines by around 30% (FilmLocal, 2025), and on reusable asset libraries to stretch a budget further. Training, in that climate, reads as discretionary. The saving from automation lands on the quarterly numbers investors watch. Training doesn’t.
The creative talent pipeline problem nobody is pricing in
Gaming has gone furthest. It has the highest AI integration in the sector and the sharpest pull-back on graduate recruitment. Since 2021 the average age of a technical hire has been climbing. That is disinvestment in junior talent, shown in a single number. The pattern is consistent. Studios ask for someone who can deliver on day one, then drop the role that would have trained that person.
The bill arrives later. Thin the junior ranks today and the senior shortage lands in five to seven years, because that is how long it takes to grow a senior. The gap between general and recent-graduate unemployment has hit record highs, so the raw talent is there. What is missing is the will to fund the years between raw and ready. It is the same dynamic behind the AI blame game, where AI takes the blame for decisions that are really about budget. The technology isn’t closing the pipeline. The accounting is.
What the salary market is telling everyone
Pay confirms the scarcity. AI-focused software engineers in the US earn around $245,000. AI engineer pay has risen from $156,000 in 2024 to over $206,000 in 2025 (Levels.fyi, 2025). Across European tech hubs, salaries are up 20% to 35% since 2023 (LiftMyCV, 2025). Remote work widens the contest, handing the strongest people the best-paid roles wherever they sit. The AI premium is real even at entry level. So a studio has a clean reason to hire a few expensive, capable people over funding an apprenticeship. That works on a spreadsheet. On a five-year view it fails, because every firm doing it is bidding for the same shrinking pool nobody is training.
The firms getting the balance right
Some hold both ends of the rope at once. WPP’s Creative Tech Apprenticeship runs a nine-month programme in creative coding, generative AI, game engines and virtual production (WPP, 2024). Most apprentices move into roles at its production arm Hogarth afterwards. The model is roughly 80% hands-on work to 20% structured learning. It builds the technical fluency and creative judgement AI can’t reproduce, and retention runs high enough that the teams are hard to copy.
That is the move. Targeted AI for the routine work, paired with a channel that keeps producing AI-native people who read a brief as well as a model. The roles coming out of this shift, creative technologists and the titles still forming around them, reward that exact blend. A lot of the strongest people moving into them are crossing in from adjacent fields like gaming rather than arriving down a graduate route.
How to walk the line
The companies that invest in people through a downturn are the ones positioned when it turns. The strongest talent can already tell which employers have worked this out and which still treat AI as a magic hire. The roles worth filling combine creative judgement with technical fluency, and the people who hold both are scarce. Cut the junior roles to protect this quarter, and you hire the shortage you made, at a premium, a few years on. Keep the rope taut at both ends, and you will still have leaders to promote when everyone else is bidding for the same scarce seniors.
