By Josh Weiss, 10to1pr

Artificial intelligence is powering a broad range of tools that can build incredible efficiencies in the construction industry. In the right hands, AI can be used to reduce costs, identify safety hazards, monitor quality control, and solve administrative scheduling issues. In the wrong hands, AI can be used to create deepfakes that can delay projects, damage reputations, destroy business partnerships, and ruin companies.

Deepfakes are highly convincing AI-generated videos, audio clips, photos, text messages, and other media designed to deceive. This may not seem like a big risk to contractors, but if a competitor is trying to make your company look bad, or an activist is trying to stop all work on a job site, deepfakes make it much easier to succeed.

Over the last year, AI technology has become accessible to anyone with a computer. The quality of deepfakes has been rapidly improving to the point where anyone with little experience can quickly create deepfakes that are so good the audience can’t tell if it’s real or fake. The ability to use this technology to manipulate and trick people for financial gain or competitive advantage has never been easier. The risks to contractors aren’t that far fetched.

• A competitor is fearful you’re going to win a contract, so right before the selection is announced, “somehow” an audio recording surfaces that sounds exactly like you and your CEO saying how you manipulated your proposal or plan to cut corners.
• A community member wants a project stopped and creates fake images showing skeletal remains on the job site, and it looks like your crews are trying to cover it up.
• A disgruntled employee manipulates a video from a job site that appears to show uncontrolled blowing dust without any mitigation efforts, simply for the goal of creating fines against your company.

All these scenarios are easy to create with AI deepfake technology, and very hard to disprove.

So, what can contractors do to mitigate risk and protect their businesses and their reputations?

Like any real-time crisis, the faster you’re aware of the problem and the risk, the better.

Consistently monitor for digital misinformation, and tell employees to report anything they hear as rumors or see online. They’re likely to learn about the deepfake attack before company leadership does, so coach them not to share any links, even if it seems like a joke.

Once you’re aware that you’re under attack, you need to immediately start your defense and put your crisis communications plan into action. Here are five key elements to handle a Deepfake crisis like a pro.

1. Identify a Third-party Expert to Say the Deepfake Isn’t Real

Deepfakes have the inherent ability to spread rapidly through social media without verification of fact. Sadly, no one is going to believe a company that simply claims a deepfake isn’t real. It is imperative to provide substantial evidence from a valid, third-party forensic expert that the video, audio, or image is fake.

While this solution sounds easy, the challenge lies in the limited availability of technology companies working on deepfake and AI detection software. Business leaders should act now to secure software partners and subject matter experts who can be ready on short notice to come to your defense. When a crisis hits, every second counts before the deepfake spreads and becomes part of the long-term narrative of your company.

2. Simplify Your Message and Act Quickly

React immediately. Most contracting companies already have a crisis playbook to guide them through incidents. During a deepfake crisis, the key elements of your crisis response plan remain unchanged, but leaders should update the plan now to include a section on deepfakes before they potentially need it.  Consider also reviewing Site Specific Safety Plans (SSSP) for communication protocols with the general contractor and other subs.

When creating talking points, start with a message of confidence. The goal of your initial statement is not to provide details of what happened or your solution, but only to clearly and simply acknowledge that you’re aware of the issue and that you’re taking it seriously to deal with it properly. After you uncover more facts about the deepfake and consider the impact it could have on projects, you can share details with stakeholders and explain how you resolved the issue. Since your initial statement is more generic, it can be written in advance as part of a crisis playbook.

3. Talk to Your Workforce

During a crisis, project managers, foremen, and site safety leaders must be ready to communicate with angry, scared workers – calming their fears and resolving their problems. Company leaders should hold small group meetings with office and field leaders to get everyone on board quickly with key messaging. The following day, incorporate it into toolbox talks and urge company leaders to visit the job site and talk with field crews to demonstrate confidence and keep morale high, while also demonstrating appreciation to the team for sticking together during a crisis.

4. Talk to stakeholders 

Depending on the severity of the deepfake, consider who else you want to tell before they find out about it themselves. This might include calling the union rep, other subs on the job site, and the general contractor.  Be honest and direct about what’s happening without speculating or guessing. Acknowledge you were hit with a deepfake attack, even if it’s just to say you’re investigating it, and don’t yet have any answers. Provide expectations and instructions on what they should do and give updates until you can resolve the issue.

5. Speak with one voice  

If a reporter contacts your company, don’t panic. Designate a single spokesperson to ensure consistency and acknowledge questions quickly – even when you don’t know the answer. If you don’t have an answer, it’s okay to admit it, and acknowledge it’s still being investigated. Later, once the facts are known or there’s more information that can be shared, do so.  As it relates to deepfakes,  you may need to initially deny its validity, but the faster that same spokesperson can show independent, third-party proof that it is, in fact, fake, the better it makes the company appear.

You’ll likely notice that several of these five tips are already part of your current crisis communications playbook.  Now, before you need it, is the time to update that script to prepare for deepfakes.

AI can be used in positive ways to help construction companies or can be manipulated to hurt companies and individuals. Our responsibility as leaders is to prepare our organizations for both.

About the Author:
As the President and Founder of 10 to 1 Public Relations, Josh Weiss has over 25 years of experience in public relations and crisis communications. He is passionate about helping companies proactively grow their brand awareness to achieve their long-term strategic goals, as well as helping companies navigate a crisis to protect and rebuild their brand reputation. He can be reached at: josh@10to1pr.com