AI Headshots for Teams: An Honest Comparison
A finance director emailed us last month asking whether she should just put her whole team through an AI headshot tool instead of booking a day. Forty people, forty seats at about thirty pounds each, done by Friday.
It's a reasonable question and it deserves a straight answer rather than a photographer being defensive. Some of these tools are good now. What follows is where they work for a company, where they don't, and how to tell which situation yours is.
If you're deciding for yourself rather than a team, our LinkedIn headshot guide covers the individual version, including when a phone photo is genuinely fine.
What are AI headshots and how do they work?
Each person uploads ten to thirty selfies, waits an hour or two, and the software returns a batch of generated images that look like them in business clothes against a studio background. Nothing is photographed. The pictures are built from patterns learned in the uploads.
Team versions add an admin layer. Someone sets up a company account, locks in approved backgrounds and outfit styles, invites everyone by email, and the tool applies the same visual treatment across all the generated sets.
That admin layer is the genuinely clever bit, and it's why the team products exist separately from the individual ones. It's also the source of most of the problems further down.
How much do AI headshots cost compared with a photographer?
Team tools sit around £30 to £45 a seat. Photographed team days in London commonly land between £40 and £90 per person at reasonable volume, falling as the headcount climbs. So the gap is narrower than people assume: at twenty or thirty people you're often comparing £35 a head against £50 a head, not against £200.
Where AI pulls decisively ahead is distance. For a team spread across six cities there's no realistic way to photograph everyone in a week, and the travel cost stops being a rounding error.
Two things get left out of that sum. You'll discard most of what the tool produces, so the usable count is well below the headline number. And somebody has to chase forty colleagues for selfies, review what comes back, and decide what to do about the six whose results are unusable. That's real time, sitting with whoever drew the short straw, and it narrows the gap again.
Are AI headshots good enough for a company website?
For some companies, yes. For any business where trust is part of what you sell, we'd say no, and the reason is consistency of a particular kind: a team page of generated portraits tends to look uniform in a way that reads as synthetic rather than professional.
The tools have improved sharply, and several now claim their output is indistinguishable from studio work. On a single portrait, viewed once, that's often close to true.
A team page is a different test. Twenty faces with the same smoothness, the same fall of light and the same absence of texture stop reading as twenty people and start reading as a set. Individually convincing, collectively obvious. It's the one context where the technology's biggest strength, applying a single treatment across everyone, works against it.
Can AI headshots keep a whole team looking consistent?
Consistency is the thing these tools do best, and it's the strongest argument for them. A photographed day gets you the same thing, whether studio or on-location, which is worth pricing before you decide. Same background, same lighting treatment, same crop across everyone, regardless of where each person was sitting when they took their selfies.
It's worth being clear that this solves a real problem. Companies with remote teams have spent years with mismatched About pages, and no amount of organising fixes it when nobody is ever in the same building.
The catch is the input. The tool applies a consistent style, but it works from whatever photos each person supplies, so the results still vary with the quality of forty different phone cameras in forty different rooms. Good selfies produce good output. Blurry ones taken under a kitchen light produce blurry output with a nice background behind it.
Can people tell when a headshot is AI?
Less often than you'd think, and that's the interesting part. A 2024 Ringover study of 1,087 recruiters found they correctly identified AI headshots only 39.5% of the time, despite 80% believing they could spot one. Skin loses its texture, hair sits too neatly, eyes go flat, and most people still can't name it.
Which makes the real finding from that study the one nobody quotes. In a blind comparison, 76.5% of those recruiters preferred the AI headshot. Once they were told it was AI, 66% were put off, and 88% said candidates should have to disclose it. The pictures are fine. What costs you is the moment somebody finds out.
Whether that matters is a question about your market rather than about photography. A software company whose customers never meet anyone can shrug at it. A law firm, a recruiter, a wealth manager or anyone whose first meeting is a video call has a harder time, because the gap between the portrait and the person is the first thing a client notices.
The awkward version of this is being asked. "Is that photo real?" is a strange conversation to have with a prospect, and there's no good answer to it.
When an AI headshot is the right call
Remote teams scattered across time zones, fast-growing companies where the headcount changes monthly, internal directories, and anyone who needs something in place this week rather than next month. In those situations the practical argument wins and we'd tell you to use a tool.
We'd also say it plainly for budget: if the choice is between generated portraits and no portraits at all, take the portraits. A team page with consistent AI images beats a team page with nothing, or worse, a mixture of holiday crops and conference badges.
When to book a photographer instead
Client-facing teams in trust-led work, leadership and board pages, anything used at print size or in press, and any company where people meet clients in person. Our guide to how to choose a headshot photographer covers how to compare options if you go that way. Those are the cases where the difference between a photograph and a generated picture earns its keep.
There's also a practical middle path that gets overlooked. Photograph the leadership team and the client-facing staff properly, and use a tool for everyone else. We'd rather a company spent sensibly on the twelve portraits that matter than thinly across ninety that don't.
Our team days run from our studio on Venn Street or at your office, priced by headcount with the per-person rate falling as the group grows. Individual sessions start from £180. If you want to sanity-check the numbers against the market, the London pricing guide has the full picture, and the team headshot day guide covers what a day actually involves.
Quick questions
Three things come up in almost every conversation we have about this: whether the tools will keep improving, whether you can mix generated and photographed portraits, and what happens to the selfies people upload. Short answers below, and we tried a few of the current tools on our own faces before writing any of this.
Won't the tools just get better?
Yes. The likeness gap narrows every few months and anyone telling you otherwise is selling something. What doesn't change is that a photograph records a real moment and a generated image doesn't, which matters exactly as much as your particular market says it does.
Can we mix AI and photographed headshots on one page?
Badly, usually. The two sit differently and the eye picks it up. If you're mixing, split by page rather than within one: leadership photographed, wider team generated.
What happens to the selfies people upload?
Worth reading the terms before you ask forty colleagues to hand over photographs of their faces. Policies vary, some are vague about retention, and facial data isn't something anyone can change later.
If you'd like to talk it through, drop us a line. Sometimes a tool will do the job, and we'd rather say so than take a booking you didn't need.