Comms AI Newsletter – #4

AI is redefining the comms playbook in a way that was unthinkable just a decade ago. The next five years will see these changes accelerate exponentially. I hope this newsletter will help colleagues navigate the disruption AI brings and make the most of some of the opportunities ahead.

Do you want to promote a job opening? Let me know and I will feature it in the next issue free of charge.

If you enjoy reading the Comms AI Newsletter, please do share with your colleagues and encourage them to subscribe. Thanks for reading.

TOP TIPS

• Creating images: Google Gemini is really user-friendly so it’s useful to create images for a campaign or collateral. You can do this directly in Google Docs or Google Slides. Be specific about the style you want (e.g. realistic, cartoon, anime, watercolour), mood (e.g. cheerful, mysterious, warm, vibrant), and colour scheme. Include the background or context in which the subject will be placed (e.g. in a studio with a white background, outdoors, or in an indoor environment). And use specific adjectives and adverbs as prompts to explain what you need.

• ‘Make AI pick a side’: AI mostly tries to produce balanced answers. It’s hard-wired in. The result can often be quite bland content. So if you make it pick a side through your prompting, you will get more interesting content and ideas back. This will help nudge the AI to produce more human perspectives. e.g. ‘Make a strong argument in favour of the Real Living Wage from a centre-left, trade union perspective and then use this to argue against it from the viewpoint of an investor or business lobby group.’ The feedback will be far more interesting than a ‘tell me the pros and cons of the Real Living Wage’.

AND NOW FOR SOMETHING COMPLETELY DIFFERENT

In issue #3 we looked at JDs for new comms roles and the lack of reference to AI skills.

Jobs and recruitment is a hot topic at the moment, and will become increasingly so as Artificial General Intelligence (AGI) takes off. The canonical definition of AGI is “a system capable of doing almost any task a human can do”. And we’re not just talking Amazon warehouse robots (which, by the way, now outnumber humans). AGI can “replace human beings in cognitively demanding jobs”. In other words, any knowledge-economy / creative role that can currently be done on a laptop is at risk.

On Ezra Klein’s podcast with Ben Buchanan, former special adviser for AI in the Biden White House, he asserts economists do not know what to do about all this. The disruption to our post-industrial way of working, about to hit labour markets in an extremely compressed time period, is incomprehensible to even the most qualified experts in the field.

Jason Cabrera posted about this too: 

“For years, we've told ourselves that AI will enhance human potential, not replace it. But that only holds true until AGI enters the equation. When a system can reason, learn, and execute tasks at superhuman speed – without ever needing rest, pay raises, or personal time – what incentive is there to maintain a traditional workforce?”

“The idea of widespread job displacement isn’t science fiction anymore. Some predict AGI could emerge by 2026, putting millions of jobs on the line – not just manual labor or routine tasks, but highly skilled professions once thought untouchable. If a single AGI agent can outperform an entire department, what happens to the employees?

“The question isn’t if change is coming. It’s how we prepare for it. If businesses, governments, and workers don’t act now, will we find ourselves reacting too late?”

AI IN THE NEWS

'Almost certain' Civil Service staff numbers will be cut in AI efficiency drive, minister says - Sky News

There is £45bn worth of productivity and efficiency savings to be made within government if AI is embraced, according to the Secretary of State for Science, Innovation and Technology Peter Kyle.

"[T]he DVLA open 45,000 envelopes every single day. HMRC is picking up the phone 100,000 times every day – this is not the way we should be doing government. This is not the way we should be running a country in the 2020s."

It comes as the Prime Minister set out proposals to recruit 2,000 tech apprentices to "turbo charge" the take-up of AI in Whitehall with the aim of modernising departmental delivery. 

Keir Starmer announced that new AI and tech teams will be sent into public sector departments "to drive improvements and efficiency in public services". He said: "No person's substantive time should be spent on a task where digital or AI can do it better, quicker and to the same high quality and standard." Downing Street said one in 10 civil servants will work in tech and digital roles within the next five years.

Google’s Gemini Robotics AI model reaches into the physical world - Wired

Google’s DeepMind has launched a new version of its AI model Gemini, fusing language, vision and physical action together to power a range of more capable, adaptive and potentially useful robots. 

AI skeptic worried about “rot-com” in the tech industry, gives robot-fried chicken a try - The New Yorker

That’s right, there are robot restaurants in LA. This was the setting for a piece with AI skeptic Ed Zitron, who claimed: “All these companies are obsessed with growth”. “But we’re approaching the end of hyper-growth in tech, and it’s making them do crazy things.” AI is just a “big, shiny bauble to say, ‘We have ideas.’ ” And investors have eaten it up.

LISTEN

Raoul Paul with Emad Mostaque: AGI, AI Agents & the Future of Work

According to Emad Mostaque, founder and CEO of Intelligent Internet, we’re now “at the point of economic social take-off… with technologies impacting real lives, real economies, real social structures.”

Two key things he references: “The magnitude of transformation with AI reaching human level in digital and physical form.” And the “diffusion” of AI into the real world. 

Emad uses the arrival of Alexa+ as an example. Alexa goes from being “stupid to smart” as it is now powered by Claude. And all of a sudden this technology is in front of every single Prime household, with millions now having access to super intelligence across the world. 

“We’ve never seen anything like it before … We are now seeing irreversible changes and nobody is prepared for it.”

Raoul points out that AI models now have an average IQ of 150 (doubling since last year). It will likely double again this year, by which case we’re then “beyond all humans”. Emad uses the chefs and the cooks analogy. We’ve gone from AI models being the “cook” that we, the knowledge controller chef, instruct. To AI being the super-chef as well as the many cooks. The question is what do we do with that level of IQ? And what do we do with ourselves?

JOBS

Director of Communications, FluidStack (New York, London, San Francisco, US Remote, UK Remote)

Global Communications Manager - EU, Securitas (UK / remote)

If you made it this far, thank you for reading and please subscribe to Comms AI Newsletter. #5 will be out soon.