Better #59

Kia ora,

Lovely to see you.

This week is all about AI, so brace yourself for some juicy insights from last edition’s AI survey.

Firstly though, it was interesting reading an interview with NZ’s Public Service Commissioner Sir Brian Roche recently, where he basically went all in on AI:

"What we need to do is have an AI world where it's the first thing. I want it to become the way we do our business, that AI is at the heart."

On one hand, I still find it pretty dumb that the starting point is literally reducing headcount rather than improving public services. But from a directional perspective, the message couldn't be clearer.

If Sir Brian were running a food company right now, he'd be saying, ‘Put protein in everything’.

If he were running a B2B software company, he'd be saying … exactly what he's currently saying 😅

Which means that being clear and deliberate around how we balance our AI use with our human capabilities, will be more important than ever.

And if our survey results tell us one thing, it’s that we haven’t yet got this balance right.

Cheers 👋
Seamus Boyer

It turns out y’all are pretty interested in AI. Or at least nosey about how everyone else is using it 👀 Probably both.

In my last newsletter I asked you to complete a quick pulse check on how comms and social media professionals are actually using AI. I didn’t know what to expect, but more than 100 of you (108 at the time of writing) took part. Thank you.

The findings show clearly that we're still very much in a messy transition period. AI use is widespread, but the norms, expectations and boundaries are still being worked out - and you are worried about that.

Which is exactly why Hayden Maskell and I have started developing a practical course and framework for communicators.

It's not ‘AI 101’ or a prompting course. We're much more interested in helping comms professionals make more strategic, evidence-backed decisions about AI - using it where it genuinely adds value, while protecting the human skills that make great communication possible: our creativity, judgement, curiosity, empathy, and critical thinking.

We'll be testing the course in person with a couple of teams next month. If you'd like your team to be considered as guinea pigs - or want to be first to hear when the course is ready - hit reply to this email and let me know.

Righto, let's get to the results.

👇

Everyone is using AI

Ok, for starters, and no real surprise here, but everyone is using AI in some way, shape or form.

About 85% of you are using it every day or several times a week, with an extra 6.5% using it occasionally.

Considering that we don’t always know we’re using AI because it’s so baked into everything these days, it may be more. But either way, it’s a lot.

In terms of perceived benefits, however, there is a real mix of opinions.

For starters, it’s worthwhile saying that nearly 10% of you don’t think it actually provides any benefit to your work at all. That's a notable minority.

Of those who do see value, 43% say it saves time on routine tasks, 14% say it helps you generate ideas, and 9% say it allows you to produce more content.

Only 5% said it improved the quality of their writing - which makes sense, given that we're the professionals in this space. But it does potentially create further downstream issues if AI is helping us produce more content without improving the quality.

Beyond those options, there were plenty of "other" responses covering more specific uses: finding things in SharePoint, sense-checking emails, working with spreadsheets, editing images etc. On the face of it, these look like genuinely good applications - very practical, and not trying to replace our thinking.

We’re winging it

This is where things get a little concerning.

On one hand, organisations are clearly signalling that AI is here to stay. About 85% of respondents said AI was either actively encouraged or accepted in their workplace, while fewer than 2% said its use was discouraged or prohibited.

What’s more, about 45% of respondents said they felt some, or strong, pressure to use AI at work.

But …

… only 16% said they have clear rules for when they should (and shouldn't) use AI. Half of respondents said they ‘mostly figure it out as they go’. Hmmm.

Training isn't filling the gap either.

Only 19% have received organisation-provided training, while 28% have learned from colleagues.

By far the biggest source of learning is trial and error. Eighty-five percent of respondents said they're essentially self-taught (NB: respondents could select multiple training sources, so these percentages don't add to 100).

There's nothing inherently wrong with learning by doing - and with AI it’s both inevitable and also positive to a degree. We’re all triallists in this global technology trial.

But when organisations are encouraging AI use, almost half of us feel pressure to use it, and most teams have few clear guardrails and no organisational training, it's easy to see why so many comms and social pros feel like they're making it up as they go.

What we’re using it for

The top 5 tasks we’re using AI for are…

1. Summarising information

2. Editing/rewriting

3. Brainstorming

4. Analysis/reporting

5. First drafts

It’s probably not surprising to know we’re using AI to summarise info. After all, we have so much to process and such limited time to do so.

But what does that mean for our level of understanding of the info?

Likewise, if we’re using AI for our brainstorming and first drafts - rather than editing or rewriting - what is this doing to the quality of our content, given that the ideas are starting, rather than ending, with AI?

Outside of these main tasks, there were plenty of quite nuanced, individual responses. Things like transcription, data processing, setting up meetings, and text reduction.

Again, these all seem pretty uncontroversial.

We’re concerned about it

We also asked whether communications professionals are becoming too reliant on AI.

Interestingly, only 31% said yes, while 42% said they were unsure.

But when we asked what worries people most about AI, a different picture emerged. The biggest concern was loss of creativity (67%), followed closely by over-reliance (56%), tied with accuracy (also 56%).

So in other words, we're not convinced we've become too reliant on AI yet - but we're clearly worried about the possibility.

There was also strong concern about losing audience trust (46%), and the written responses reinforced that theme again and again. People talked about:

• "Loss of uniqueness."

• "Sounding generic and inauthentic."

• "Loss of human voice."

• "It's so bloody boring reading AI-generated content."

It sounds (and feels) like we are very worried about this. So it’s probably good we have this discussion now, while we're still shaping our habits, rather than a few years down the track when those habits are much harder to change.

How people are getting it wrong

Finally, we asked people to complete the sentence: "the biggest mistake communications professionals are making with AI is..."

And boy did you complete that sentence.

We got 100 written responses - which again says a lot about how you feel around this.

The dominant theme was that communicators are outsourcing the wrong things to AI - asking it to do the writing, the creative thinking, and even the strategic thinking, rather than using it as a tool to support those activities.

Many respondents also worried that people are accepting AI output at face value, publishing it with little editing or critical thought.

Which is resulting in too much content that feels generic, predictable, and just really, really boring.

Another common theme was overestimating AI's capabilities - treating it as the answer to every problem instead of using it deliberately for the tasks where it genuinely adds value.

Interestingly, respondents also saw the opposite extreme as a mistake. A number argued that refusing to engage with AI at all is becoming increasingly risky, particularly as organisations begin expecting communicators to use it as part of their everyday work.

In other words, the conversation is all about how we use AI, so that we keep the crucial human element in our comms and content. We’ve almost overwhelmingly accepted that AI is part of the future of our work, but we’re still not sure exactly what part.

There were also a few cheeky responses that deserved a mention:

• "Em dash overuse."

• "Thinking it makes them a good designer."

• "Forgetting to use their mind muscle first."

All I can say is: agreed 😆

Ok, that's it for this week - I hope you found the results as interesting as I did.

If you'd like to join the wait-list for Hayden and my upcoming AI course for communicators, just hit reply and let me know. We'll share more once we've tested and refined it.

I'm also looking to take on one or two new strategy clients this quarter. If your organisation is struggling to earn attention, build trust, or get more impact from its communications, I'd love to have a chat.

And I’ll see you again soon for more tips on how to get just a little bit Better 👋