This article is an on-site version of our Working It newsletter. Subscribers can sign up here to get the newsletter delivered every Wednesday. Explore all of our newsletters here

Hello and welcome to Working It.

We are making a second season of FT Working It films, and the first topic will be the meteoric rise of AI agents. It’s certainly timely: OpenAI’s ChatGPT launched its agent this week, while SoftBank is aiming to create 1bn types of AI agents by the end of 2025, a scenario in which “AI agents proliferate by themselves”. I find this hard to process 🫣.

More cheerily, we filmed last week at the Paris HQ of recruitment platform Welcome to the Jungle, where the meeting rooms have Star Wars-universe names because WTTJ’s founder and CEO, Jérémy Clédat, is a fan. Does anyone else have cool meeting room names? (I fear the FT’s on-brand “currency” rooms — Peseta, Dollar, Mark etc — won’t win this game 😂.)

Interplanetary Teams meetings

Email me about agents, sci-fi, meeting rooms, or anything vaguely work-related: isabel.berwick@ft.com.

Read on for a surprising “key takeaway” from that viral CEO/HR/Coldplay clip (come on, did you think I wasn’t going to mention it?) and, in an attempt to channel Kim Scott, I hand over Office Therapy for reader feedback 👀 about last week’s problem involving a clock-watching colleague.

That Coldplay concert clip showed us another problem hiding in plain sight 🤯

Once we’d realised that now-former Astronomer chief executive Andy Byron was not an actual astronomer (or astronaut), some of us turned to the meaty question of what, exactly, Astronomer does. The answer is data: according to a handy Wall Street Journal post-Coldplay primer, it provides “the infrastructure that helps [companies] integrate their data into the applications where they need to use it. The company said it has worked with Apple, Ford and Uber.”

Astronomer is among many companies offering solutions to the problem of how to manage, and make the most of, the data held inside every organisation. And that data quality is foundational to any AI rollout. Without good, accurate data, that is also well-organised, unbiased, un-siloed, and optimised in the right ways, a rollout will probably fail to get past a basic level. (“Treat data as a product” has become a buzzword for a reason: it’s true.)

I had not previously considered this data problem with any rigour, despite writing and talking (a lot) about AI at work. It took that viral clip to make me go deeper.

It’s hard to get an idea of the financial damage done by bad data. A frequently cited HBR report, for example, offers a figure of $3tn a year as the cost of bad data to US business — but that’s from 2016*. We need better figures, for sure, but many people believe that organisations just need to sort their data out before spending vast amounts of money on AI.

I asked Euan Blair about the data conundrum when I interviewed him in London this week. Euan is founder and chief executive of Multiverse, a digital skills and apprenticeship platform. Helping clients to build and deliver AI training programmes for their staff is a big part of what Multiverse does. Does the underlying data problem come up a lot? Oh yes. The company often spends time on this with clients, before they get to the AI skills part. Euan told me they work “on basic data literacy, data analysis and trying to democratise access to data away from small data teams who will frequently be overwhelmed with the amount of requests they are getting”.

And he highlighted the similarities between how we should be working with data and AI: “Increasingly, people don’t need ‘specialist’ AI individuals working with them, they can start to build themselves, and they can become creators. With data, everyone should be able to self-serve, otherwise companies just become overwhelmed.”

At Atheni, which has built a tool (“like a satnav”) to guide leaders and teams as they use AI, co-founder Mackenzie Howe told me: “We’re seeing two completely different approaches right now. Some organisations are powering straight into AI, learning and troubleshooting as they go and taking a very agile approach. Others — and we spoke to one just last week — said they had the idea they shouldn’t embark on any AI projects until their data is 100 per cent perfect.”

Mackenzie’s pragmatic take is that perfect data doesn’t exist, and leaders can start working immediately with what they have. “Waiting for perfect data before starting AI is like waiting for traffic to clear before learning to drive. You’ll be waiting forever while everyone else gets where they’re going.”

I am never going to take data for granted again.

(Also: maybe don’t take an office affair partner to a stadium gig 🧐. That data point is well and truly unleashed.)

*Full disclosure: I use the FT’s ‘walled’ ChatGPT and NotebookLM for research. It’s saving time and throwing up new sources — but I double check it all. And the writing is all me ✍🏻.

  • In a nutshell: Good data is foundational for AI transformation, so don’t go all-in on AI investment without it. Equally, don’t miss the boat. Delaying all AI rollouts may mean staff miss out on an important learning curve.

  • Want more? Useful insights in Gartner’s reports, for example “What Is AI-Ready Data? And How to Get Yours There”. And Multiverse’s recent Skills Intelligence Report highlights data and AI skill gaps in the UK workforce, and how to fix them.

Office Therapy

Last week’s problem covered the matter of a clock-watching colleague. To recap, a reader is annoyed by a co-worker who “arrives at work at 9.30am and stands up at 5.28pm, two minutes for packing up, and out at 5.30pm. At lunchtime they take a full hour.” Commenters on FT.com had little sympathy for this dilemma. “The writer needs to pop to Homebase, buy a step ladder and use it to get over themselves.” 😱

Working It reader Trond Stroemme, meanwhile, wrote in with an interesting cultural perspective: “In my native Norway, this is the norm, for a very logical reason. Norway is a country where life/work balance is sacred. Not necessarily because people don’t want to ‘put in the extra effort’, but because, quite simply, when you have kids at school or in kindergarten, you need to pick them up punctually. And, of course, a lot of people are parents.

“Norwegians wouldn’t come to the office at 9.30, however, more like 8.00. Which means they leave at 4pm, promptly. Lunch break, on the other hand, is only 30 minutes. Childcare is abundant in Norway, at state-subsidised prices, meaning the idea of au pairs or nannies simply does not exist. You leave home at 7.30am, deposit your kids at the kindergarten, and are normally in the office by 8am. It’s not at all uncommon to see your boss leave before you do, in the afternoon. Maybe she (or he) simply has longer to travel to pick their kids up.”

Email your problems and dilemmas to isabel.berwick@ft.com. We anonymise everything, properly.

Five top stories from the world of work

  1. You can’t be at the peak forever: how CEOs are learning to pace themselves: The stresses of the job are nothing new, but always-on culture is taking a toll on leaders. Anjli Raval talks to the doctors and coaches helping them to manage what can be a long-haul endurance test.

  2. A fresh phone hell in the office: Younger staff may dislike making telephone calls, but Pilita Clark finds that they may also answer with a silence. It’s a generational shift, for sure, but one that’s not right for an office environment.

  3. AI is reinforcing the dominance of English in the workplace: English was already the global language of business even before AI arrived, although fluent Dutch speaker Simon Kuper outlines the benefits of speaking a minority language.

  4. The rise and fall of Fred Goodwin: The playwright James Graham is tackling the saga of the rise and fall of the Royal Bank of Scotland in his new play, Make it Happen. Lionel Barber, FT editor at the time of the financial crash, writes compellingly about the story — and his own impressions of Goodwin.

  5. Corporate HQs are becoming more transient: A surprising number of businesses move their head office, as Oliver Ralph outlines. Prudential, for example, has two head offices, in London and Hong Kong, but he notes that its “centre of gravity” is now in Asia.

One more thing . . . 

In a recent Working It callout for reading/watching/listening recommendations for the summer 😎, I said there wasn’t an “it book” for 2025. Not so, says Claire Jermany Grange, senior manager for financial communications at Aviva: “There 100% is an ‘it’ novel this summer — it’s fairly heavy but an incredible read: The Names by Florence Knapp. Highly recommend!” I can’t find an FT review, but we did discuss the audiobook. (Fellow audiobook lovers, do look out for these columns by Alex Clark).

Lorrie Fay, media communications manager at the Royal Academy of Engineering, suggests two TV series: “I’m loving Karen Pirie (ITVX) based on the work of Scottish crime writer, Val McDermid. The cold cases are set in the 1980s and Karen is working in modern times. The second series is out now. And Bookish (U&Alibi) is a crime drama, set in 1940s postwar London. It’s created by Mark Gatiss and is centred around a lavender marriage.” I’ve seen both of these and endorse Lorrie’s recs 💯.

Please keep them coming 📝.

A view from the Working It community 📸

A mindfulness moment this week (sorely needed), from Dorina Timbur in Leeds. She writes: “I am a technical seller and mainly work from home. The past few days my walk breaks have taken on a new meaning as I started visiting and feeding a nearby family of wild geese. All the urgency and AI doom in the workplace just take a back seat while I look at these beautiful babies.” Mood 😌.

Honking lovely

Send us your workplace or work from anywhere views, wildlife or pets optional: isabel.berwick@ft.com. Dorina (and anyone whose photo is printed) will receive a “lucky dip” of workplace and management books.

And finally . . . festival time 🎪

The FT Weekend Festival returns to Kenwood House on September 6, with a stellar line-up of speakers, including Yuval Noah Harari, Lyse Doucet, Sharon White, Nick Clegg and David Nicholls, as well as many FT journalists. As a newsletter subscriber, save £25 on your ticket with promo code FTxNewsletter25. I hope to see some of you there 👋.

Recommended newsletters

One Must-Read — Remarkable journalism you won’t want to miss. Sign up here

Sort Your Financial Life Out — Learn how to make smarter money decisions and supercharge your personal finances with Claer Barrett. Sign up here