I have several years of experience directing public information for state and national criminal justice agencies, including law enforcement agencies. The Maryland Emergency Management Agency was part of the Maryland Department of Public Safety when I was their director of public information.
The first thing you need to understand is that I or anyone else can create green screen videos using a very realistic digital television news set, import commercially available videos, and spread disinformation in about twenty minutes. It’s easier to create fake audio. Anyone can take AI-generated photos, but the easily accessible commercial tools can do it better. They can be sent to social media via the dark web from platforms and IP addresses around the world.
So understanding disinformation means recognizing that it can be created faster and better than anyone can imagine. It’s equally important to understand that if I can do it, what happens when cartels with endless budgets and the dark web engage in the same process?
What happens when hundreds of thousands of Phoenix-area residents are told their water has been deliberately contaminated? What happens when the same campaign tells El Pasoans that their food supply has been compromised? All you have to do is use chemicals in small areas to create real-world examples that make the disinformation campaign more credible.
The cartels have access to a large number of armed drones that they use against each other and against Mexican law enforcement. It’s not outside the realm of possibility that they could be used against American cities.
Yes, AI-generated media makes it more difficult for the government to determine the validity of all media posted online, but every state has the capacity (or should have the capacity) to accurately measure what is being offered through social media and respond with corrected information.
The main question is whether law enforcement or government has the capacity to respond hundreds of false messages on social media in a short time.
The problem isn’t AI, although that makes the mission of providing accurate information more challenging.
There have always been obstacles to providing accurate information to the public. Photoshop images, the ability to purchase commercially available video footage, and the use of green screen technology have been around for decades.
For example, if the problem is the stability of flood waters breaching a dam, sending a video simulating a dam burst via social media could cause massive panic.
I have the tools to create a very realistic video using a news set of footage readily available on commercial sites. I can cause enormous panic.
The topic or situation doesn’t matter; if I want to cause unrest or stir up hysteria, I can do that.
The real problem is rumor control
The response is not primarily digital. It requires an enormous amount of personal strength. How do you ensure that the dam does not break? Make sure there is someone stationed at the dam with the authority to communicate with management. That requires a lot of planning and coordination.
This is why the public affairs department of any police, criminal justice or emergency management organization must be prepared. It requires a huge effort from public information officers from various agencies to address, validate or dispel rumors, fake videos, audio or fake photos. This requires experts from different agencies. Each state must quickly gather a large number of people, in twelve-hour shifts. They must have the means to communicate.
As director of public information for the Maryland Department of Public Safety and the Maryland Emergency Management Agency, we called FEMA to evaluate training sessions focused on nuclear power plant meltdowns or chemical weapons exercises to ensure the appropriate information, equipment and personnel were available.
But the event didn’t matter; the state offered its resources to each requesting agency, and I then called on about thirty experts from various agencies to serve as my rumor control team. This had to work 24 hours a day until the event was over.
There was one person responsible whose main task was to insist on the verification of all information.
We monitored the news media and social media for signs of false information or suspicious media releases.
Once verified, I, as the primary spokesperson, would release that information to the public after consultation with the concerned agency. There was one spokesperson, me. Multiple spokespersons would inevitably provide incorrect information.
We would also send public information officers or agency personnel to the scene of an emergency so that we would have people who could verify information immediately.
Again, there is a process in every state to identify false or misleading information. Yes, AI makes it more challenging, but we have trained people to verify fake images, misleading videos or audio.
The media becomes your partner. You can’t run an anti-disinformation team without them. They need to know what damage is caused by the release of suspicious material. They must be willing to come to you first before releasing any information. Building good media relations in advance is a wise move.
The media MUST have quick access to your spokesperson. There MUST be people assigned to the primary spokesperson to catalog and prioritize incoming media calls. If you use the Associated Press as your first response mechanism, it will immediately be passed on to everyone else.
Conclusion
Few in the public understand what states do to verify information and the process of getting facts out. We know the reporters and how to reach them through the Associated Press or direct contacts.
Every agency in the country has access to the state emergency management function and the resources available for it.
In Maryland, we trained the team twice a year, and the key spokespersons were often FEMA trained. Within my department, we have trained our part-time spokespersons to handle emergencies within our twelve departments, but we have also used our part-time PIOs to assist other agencies.
It’s all part of a very precise plan to put the right people in the right places so they can quickly make decisions that are in the best interest of the public. It is more than possible to spot counterfeit AI-generated products, mainly through our network of experts or by having people stationed on site.
The video or photo of a large dam bursting won’t cause lasting damage if you have staff who can verify it is fake.
The same goes for AI versions of all media.
