AI-generated submissions: How to detect, prevent and manage them fairly

As a content writer, I notice almost daily how profoundly artificial intelligence has changed the way we write. Certain sentence patterns, verbose copywriting or the overuse of em dashes have become quiet signals that a machine had a hand in creating the work.

Every industry is experiencing some form of this. And that includes awards management, where awards managers strive to recognise original work and true excellence. Nobody wants to give recognition to someone who did not conceive and create the work themselves.

Fairness has always been central to awards programs, but with AI tools freely available to everyone, the issue has taken on a new urgency. How do you spot AI-generated content in submissions today? And how do you handle it without treating every participant as a suspect?

When technology moves faster than your processes

Awards programs run on trust.  Program managers and judges rely on submissions to genuinely reflect the work and achievement of participants. That trust is quickly undermined when large language models can produce convincing, well-structured entries in seconds.

The problem is not necessarily the technology itself. Many participants use AI tools legitimately — for spell-checking or translation, for example. The difficulty arises when entire submissions are generated with no real substance behind them. Program managers and judges now face the challenge of identifying manipulation without burdening the participant experience with excessive suspicion.

There is another complication: AI-generated text is getting better and harder to detect. Instinct alone is no longer enough.

Clear guidelines create confidence for everyone

As uncomfortable as the topic of AI may feel, it also presents an opportunity. Programs that lay out clear and transparent rules around AI  build trust with participants, judges and sponsors alike. A holistic approach to AI-generated content sends a clear message that your program takes integrity seriously.

A well-rounded strategy for preventing and detecting AI-generated content combines clear rules, technical support and well-trained judges.

Data security also plays a role here. The moment you review submissions or request evidence of authorship, you are handling sensitive data. A platform that treats privacy and configurability as core considerations from the outset makes this process significantly easier.

Award Force, for example, is built to allow program managers to configure the review process with flexibility, without compromising on security or privacy.

Practical steps for addressing AI-generated content in submissions

How do you actually check whether content is AI-generated, and what do you do with the results?

1. Set clear guidelines upfront. Define your policy around AI usage. If you do allow, how and with what tools? Participants can only follow rules they know about. The industry is already moving in this direction: according to a recent Awards Trust Mark survey, 43 percent of awards organisations polled already had a clear policy on AI-generated content in 2025, compared with just 13 percent the previous year.

2. Train your judges. Typical indicators of AI-generated text include repetitive sentence structures, points listed in groups of three, or a certain surface-level polish that lacks real depth. Overuse of particular stylistic devices, such as em dashes or phrases like “it’s not X, it’s Y”, can also be signals.

That said, a single characteristic is not evidence. Closer scrutiny is only warranted when multiple indicators appear together. After all, AI language models learned their style from humans. Even in my own writing, I tend to list things in threes and occasionally reach for an em dash to emphasise a point.

3. Use supporting technology wisely. Use AI to detect AI. There are tools that can assess whether content is likely to have been AI-generated. These tools return probabilities, not proof, so do not use them as the sole basis for a decision. AI tools, on the management side, can be used intentionally to protect your awards program and detect plagiarism.

4. Invite clarification rather than rushing to disqualify. When a submission raises concerns, give participants the opportunity to explain or provide evidence of their work. This protects against misjudgements and strengthens the credibility of your evaluation process at the same time.

5. Document your decisions. Record the reasons why a submission was reviewed or rejected. This creates a clear audit trail if questions arise and helps you refine your criteria from one cycle to the next.

6. Configure your platform to match your requirements. Every awards program has different needs when it comes to review processes, approvals and data privacy. A configurable submissions platform lets you build the workflows you need without losing control of sensitive data.

Genuine achievement deserves recognition

AI-generated submissions will continue to be a challenge for awards programs, and the technology shows every sign of advancing rapidly. Program managers who establish clear guidelines now, train their judges and invest in reliable, well-configurable platforms will be well placed to respond.

Ultimately, it comes down to a simple question: how do we make sure recognition reaches the people behind genuine achievement? Award Force supports awards programs around the world in answering exactly that question every day, with solutions that put security, privacy and fairness at the centre.