Hi Reader,
The EU started labeling AI content this week.
As of August 2, chatbots have to say they are chatbots. AI-generated images, audio, and video have to be marked. Penalties run up to 3% of global revenue.
LinkedIn spent the same seven days building a button so users can report posts that feel synthetic. Snapchat stopped recommending fully AI-generated video. The whole internet suddenly got very interested in telling you what is real.
Which makes a strange backdrop for the one thing most comms teams still cannot do.
If a convincing video of your CEO showed up tomorrow, saying something they never said, who verifies it? How quickly? And what do you put out while you are still figuring that out?
Most teams have a crisis plan. Almost none of them have an answer to that.
ππΎ Here's the prompt:
π― Expert Mode: The "Deepfake Dry Run"
Scenario: A synthetic video of your CEO is circulating. It is good. A reporter has already emailed. You have 60 minutes today to build the muscle you will need on the day you actually need it.
Instructions:
1οΈβ£ Pick your most exposed principal. Usually the CEO, sometimes the founder, occasionally the scientist who does all the podcasts. Write down three things: how much public video and audio of them exists, who inside your company could technically confirm or deny a clip, and who has authority to approve a statement at 11pm on a Saturday.
2οΈβ£ Write one paragraph describing the fake. Be specific. What does the clip show, where did it surface first, and who is amplifying it. Vague scenarios produce vague plans.
3οΈβ£ Paste both into Claude or ChatGPT with the prompt below.
Act as a crisis communications advisor running a tabletop exercise. Below is a scenario in which a convincing synthetic video of our senior executive is circulating, along with an honest inventory of our current verification capability and approval chain.
β
Produce five things. First, a 90-minute response timeline broken into 15-minute blocks, starting from the moment we become aware. Second, a verification chain with a named role responsible at each step and a realistic time estimate for each step. Third, a holding statement we could send in the first 30 minutes that does not confirm or deny anything we cannot yet verify. Fourth, a platform escalation and takedown sequence in priority order. Fifth, the three questions a reporter will ask that we currently cannot answer.
β
Work only from the capabilities I actually listed. Do not assume we have forensic tools, a standing legal contact, or 24-hour monitoring unless I said so. Where a step depends on something we do not have, say so plainly and tell me what the workaround costs us in minutes.
4οΈβ£ Read the verification chain with a stopwatch mentality. Add up the realistic time estimates. If the total is longer than the time it takes a story to publish, that number is your actual finding, not the plan.
5οΈβ£ Save the holding statement somewhere your on-call person can reach it without a laptop. Then put a 20-minute re-run on the calendar for next quarter with a different scenario.
Suggested Tool: Claude (best at refusing to assume capabilities you did not give it, which is the entire point of this exercise)
Why You'll Love It
- It turns an abstract fear into a timeline with names on it, which is the only version of a crisis plan that survives contact with a real crisis.
- It exposes your verification gap in an hour instead of during the incident, and that gap is almost always the real story.
- It gives you a holding statement written by a calm person, because the one written at 11pm by a panicked person is the one that gets screenshotted.
Every comms team has run a media training. Very few have run a verification drill. The next 12 months are going to make that the more important of the two.
Talk soon,
Chris
β‘ The Cowork Version
Same drill, except Claude reads your actual crisis plan and org chart, finds the gaps itself, and leaves a finished tabletop packet in your folder.
Setup (one time):
- Drop your current crisis communications plan into a folder. A PDF, a Word doc, or a set of Google Doc exports all work.
- Add an org chart or a simple contact list showing who approves external statements and who is reachable after hours.
- Add anything you have on your principal's public footprint: recent interviews, keynote videos, podcast appearances, or just a list of links.
- Optional: connect Slack or email so Claude can check whether your named approvers are actually the current ones.
Run this Cowork command:
Run a deepfake tabletop exercise using the files in this folder.
β
1οΈβ£ Read the crisis plan and the contact list. Build an honest inventory of what we can verify today, who can authorize a statement, and how long each of those steps realistically takes after business hours.
β
2οΈβ£ Read the public footprint material and assess how much source audio and video exists for our principal. Note which formats would be easiest to convincingly fake.
β
3οΈβ£ Write three distinct scenarios at increasing severity. A fake clip on a fringe platform, a fake clip picked up by a trade outlet, and a fake clip a national reporter is calling about.
β
4οΈβ£ For each scenario, produce a 90-minute timeline in 15-minute blocks, a verification chain with named owners from our actual contact list, and a holding statement.
β
5οΈβ£ Flag every point where the plan depends on a capability we do not have in the source files. Do not invent tools, vendors, or roles that are not documented. Rank those gaps by how much time each one costs us.
β
6οΈβ£ Save the three scenarios as one tabletop packet, the holding statements as a separate one-page card, and the gap list as a short brief I can take to leadership.
What you'll get:
- A three-scenario tabletop packet with real names from your org chart, not placeholder roles
- A verification chain with time estimates you can add up and argue about
- A one-page holding statement card your on-call person can read on a phone
- A ranked gap list showing exactly what your current plan cannot do, priced in minutes
- A leadership brief that makes the case for whatever you are missing, without you having to write it
Want it on autopilot? Ask Claude to save this as a skill and re-run it every quarter against your updated crisis plan. The scenarios change, the gap list gets shorter, and you have a dated paper trail showing the team practiced.
Here are a few articles from the past week to help keep you up to speed:
1. EU's new AI disclosure rules: how should PR adapt?β
PRWeek UK
August 3, 2026
Disclosure obligations under the EU AI Act came into force on August 2nd, and PRWeek asks what that actually means for agencies running AI-enabled campaigns. The detail with teeth: liability sits with whoever deploys the AI, not the vendor who built the tool; it applies to any content reaching an EU audience, and fines run to 15 million euros or 3% of global turnover. The sharpest practical note comes from We Are Warriors founder Effie Kanyua, who points out that a label applied at the point of creation is long gone three shares into a WhatsApp group, so your disclosure has to survive the content's journey, not just its launch.
2. The Scoop: OpenAI responds to backlash over luxury influencer tripβ
PR Daily
August 5, 2026
OpenAI's "Summer Club" creator retreat read as a luxury brand trip instead of an education play, and the content backfired. The company spent the week explaining an event it had designed to explain itself. A useful reminder that the format of a thing carries more meaning than the intent behind it, and that creator programs get judged on what they look like, not on what the brief said they were for.
3. Social media updates and new features to know this weekβ
PR Daily
August 4, 2026
Pangram estimates 41% of long-form LinkedIn posts show signs of AI use, the highest of any platform studied. LinkedIn is now testing a "seems like AI slop" report button, will privately flag creators whose posts read as inauthentic, and is retiring its own Enhance Your Post tool. Every executive visibility program running on ghostwritten, AI-assisted posts now has a distribution risk attached to it.
4. The Loudest Voice in the Room is Losing the Argumentβ
O'Dwyer's
August 5, 2026
Evan Nierman argues the chase for trophy tier-one placements is over, and backs it with a Muck Rack review of more than a million AI-cited links showing 84% trace back to earned media, with engines reaching for niche outlets on industry-specific queries. One well-placed niche story persuades a human today and trains a model about you tomorrow. That is a sharper argument for earned media budget than anything currently on your one-sheet.
5. A record-breaking eight Pulitzer awardees disclosed AI use this yearβ
Nieman Lab
August 3, 2026
Five winners and three finalists itemized their AI use to the judging committee, a record. Journalism's highest award now treats AI assistance as normal and disclosure as table stakes. If Pulitzer winners are writing down exactly how they used it, "we don't really have a policy on that" gets harder to say out loud to a client or a reporter.
β