Ship It or Skip It

One star.

Imagine getting that on a game you just released after countless days and nights of work. Not because the game is bad. Because you put AI in there hoping to cut a few hours on art and players came after it with pitchfork reviews. The AI debate in game dev is brutal right now. Think twice before that choice costs you.

Five questions. One honest answer about whether AI is the right call for this specific task, right now, before you find out the hard way.

Question 1 of 5
What type of task is this?
Select a task type to continue.
Question 2 of 5
Will players see, hear, or interact with this output directly?
Select an option to continue.
Question 3 of 5
Select an option to continue.
Question 4 of 5
Does your team have someone who has built a repeatable process around this specific AI tool?

Not just opened it once. Actually shipped things with it across multiple real projects.

Select an option to continue.
Question 5 of 5
Select an option to continue.
Risk score
The environment you are operating in
The sentiment from players and developers right now is: we are watching, and we will call it out if you do not do this right. Even though nobody has fully defined what right is yet. That is the actual environment you are operating in.
Free Download
Studio AI Use Policy Template
A baseline policy covering approval tiers, QA verification, platform disclosure, voice talent requirements, and recordkeeping. Built from published studio policies, SAG-AFTRA 2025 IMA provisions, and legal guidance. Fill in your studio name and route for sign-off.
The room you are walking into

Game development has seen historic layoffs. A lot of developers watched colleagues get walked out while executives talked about AI efficiency in the same breath. So when a vendor shows up with a demo and the word "game changer" in their pitch deck, the room is not exactly neutral. Just so you know.

What is actually running through the developer's head during your demo

While you are showing them the impressive output, they are running a completely different calculation.

Who on my team absorbs the time to learn this? We are already behind on two milestones.
What happens to my current pipeline when this gets dropped into it? We have tools that talk to each other in very specific ways and one new thing can break three others.
What does this look like when it fails in production at 11pm before a cert submission? Because it will fail. Every tool fails. What does that cost us?
If this tool improves our output, does that mean we need fewer people? Am I making the case for my own team to get smaller?
If I recommend this and it goes badly, that is on me. If I push back and it turns out to be useful, that is also on me. Either way I am carrying the risk.

None of these get answered in a demo. Most do not get asked out loud. All of them are in the room.

What a demo does not show

A demo shows a developer typing a perfect prompt into a tool and getting something impressive back in four seconds. It does not show what happens when the input is a 200 asset project with three years of technical debt, a cert submission in 48 hours, and the one person who knew how this pipeline worked just got laid off last quarter.

The vendors who get the furthest in a room full of game developers are the ones who make it clear, from the first sentence, that the human developer is the most important asset on the team. Not the AI. Not the tool. The person sitting across from you who has been doing this for years and is now being asked to evaluate whether your product is going to make their job better or their team smaller.

Saw this firsthand at a Runway summit. When a vendor leads from that position the room responds completely differently. It is not subtle.

Everything else follows from that.

1. Lead with their problem, not your tool

If you genuinely believe the developer matters most, you have done the work to understand what they are actually struggling with before you walked in. Open there. Not with a feature list.

2. Drop "game changer" from your vocabulary

Developers have heard it so many times it now triggers the same response as a car alarm. Nobody looks up. If your tool is genuinely good, the work shows it. The word does not.

3. Show what happens when it breaks

Your demo worked because you controlled everything. A developer watching it is already imagining what happens when the input is three years of technical debt and there is a cert submission in 48 hours. Show that version. The vendor who is honest about failure is the one who gets trusted.

4. Bring a real example or bring nothing

A studio, a real deadline, what went wrong, how it got fixed. That is worth more than any benchmark slide. Real developers made it work in a real pipeline under real pressure. That is what the person across from you needs to see.

5. The layoffs are in the room whether you acknowledge them or not

Developers have watched colleagues lose jobs while AI gets credited in the same breath. You do not have to address it directly. But a vendor who acts like the resistance is irrational, rather than completely understandable given the last two years, will lose the room before slide three.

6. If your tool costs something to adopt, say so upfront

Pipeline disruption, learning curve, the weeks it takes to build a real process around something that looked effortless in a demo. If you respect the developer, you respect their time enough to be honest about that cost. Then talk about what you are offering to offset it. Because they are already doing that math in their head while you are still talking.

You are not anti-AI because you asked hard questions. You are doing your job.

You are allowed to ask what the integration actually costs. You are allowed to say the demo does not reflect your pipeline. You are allowed to ask what happens when it breaks at 11pm the night before cert and whether the vendor has a support line that answers at 11pm.

You do not have to frame your hesitation as anything other than what it is. You have a game to ship. That is the priority. Use the Decision tab to put some structure around the conversation.

The current moment
Nobody fully knows where this lands yet. The vendors who admit that tend to get further than the ones still saying "game changer."