IBM just paid $17M over DEI. This isn’t about the money — it’s about the signal. ⚖️

TL;DR: The US government is now actively targeting corporate DEI programmes — and Big Tech is quietly recalibrating what “diversity” looks like in practice. 🧠

IBM has agreed to pay $17 million to settle a US government probe into its diversity, equity, and inclusion (DEI) practices.

Importantly, IBM didn’t admit wrongdoing. The government didn’t back down either.

So everyone walks away saying:
👉 “we’re right”
👉 but with the DOJ also collecting $17M

A very modern compromise. 💼

The case sits inside something bigger.

The US Department of Justice is now using the False Claims Act — yes, the Civil War-era fraud law — to go after DEI programmes, particularly for companies with federal contracts.

The allegation against IBM?

That it tied hiring, promotions, or bonuses to demographic targets — effectively turning diversity goals into something that could be interpreted as discriminatory.

The government’s framing is blunt: DEI, if misapplied, becomes illegal bias.

Which is… a pretty fundamental shift in tone. 🚨

Zoom out, and this is less about IBM and more about direction of travel.

Over the past year, US policy has moved aggressively against DEI across:

• federal agencies
• universities
• and now private contractors

Many companies have already started quietly:
👉 scaling back programmes
👉 renaming initiatives
👉 or removing targets altogether

Not because they’ve changed their views — but because the legal risk has changed.

And here’s where the irony kicks in.

For years, companies were under pressure to:
👉 measure diversity
👉 set targets
👉 show progress

Now they’re being told those same mechanisms… might get them sued.

Same data. Completely different interpretation. 🧠

For UK readers, this matters more than it looks.

Because while this is a US case, large tech companies operate globally. And when compliance risk shifts in the US, it tends to ripple outward into:

• global hiring policies
• reporting standards
• and how companies talk about diversity at all

In other words:

The loudest debates might be happening in Washington —
but the policy changes won’t stay there. 🌍

So what?🧠

For companies: DEI is moving from a branding exercise to a legal risk surface. Expect fewer public targets, more vague language, and a lot more lawyers in the room.

For operators and leaders: the question is no longer just “how do we improve representation?” — it’s “how do we do that without triggering regulatory scrutiny?”

For early-career talent and underrepresented groups:

This is the uncomfortable bit.

Progress that relied on formal programmes may slow… just as structural inequalities haven’t gone anywhere.

And for the industry overall:

We’re entering a phase where inclusion, compliance, and politics are colliding — and no one has quite figured out the stable middle ground yet.

Read more:
https://www.reuters.com/world/ibm-pay-17-million-settle-discrimination-allegations-doj-says-2026-04-10/

The UK is finally calling racism in maternity care what it is. Now comes the harder part: fixing it. 🏥

TL;DR: New anti-racism principles for midwifery training aim to tackle the Black maternal health crisis — but changing outcomes means changing systems, not just curriculums. 🧠

The Nursing and Midwifery Council has announced new anti-racism principles for midwifery education across the UK — explicitly aimed at tackling one of the most uncomfortable truths in healthcare:

Black women are three times more likely to die during pregnancy or shortly after childbirth.

Let that land for a second.

This isn’t a marginal gap. It’s a structural one. 📊

The new plan is to embed anti-racism, bias awareness, and cultural competence into how midwives are trained — shaping the next generation of practitioners before they ever step onto a ward.

On paper, it makes sense.

If bias exists in care, address it in training.

If outcomes are unequal, change how professionals are prepared to deliver care.

Simple. Logical. Necessary. 👍

But here’s where it gets more complicated.

Because the data has been clear for years.

Reports from MBRRACE-UK and others have repeatedly shown:

• Black women face significantly higher maternal mortality
• Asian women also face elevated risk
• Black babies are over twice as likely to be stillborn

This isn’t new information.

So the obvious question becomes:

Why has it taken this long to move from evidence to intervention? 🤔

There’s also a deeper tension running through this.

Training matters. But outcomes in healthcare aren’t just shaped by individual behaviour. They’re shaped by:

• staffing pressures
• resource constraints
• systemic bias
• and how institutions are designed

In other words:

You can train better midwives…

…but if the system they enter hasn’t changed, progress may be slower than headlines suggest. 🧩

And then there’s the broader context.

At a time when DEI efforts are being rolled back or challenged in other sectors, the UK healthcare system is doing the opposite — making anti-racism more explicit, more formalised, and more embedded.

Two very different directions.

Same global conversation. 🌍

So what?

For healthcare: this is a meaningful step — but only one piece of a much larger puzzle. Training can shape behaviour, but it can’t fix structural issues on its own.

For policymakers: the real test isn’t announcing principles — it’s whether outcomes actually shift in the data over time.

For the wider tech and business world: this is a reminder that inequality isn’t abstract. It shows up in life-and-death outcomes, not just hiring stats or pay gaps.

And for everyone else:

If a system can produce a 3x mortality gap…

…it’s not a knowledge problem.

It’s a design problem. 🧠

Read more:
https://www.nmc.org.uk/news/news-and-updates/new-nmc-anti-racism-principles-will-support-universities-to-tackle-black-maternal-health-crisis/

Western carmakers are slowing down on EVs. China is not waiting. 🚗

TL;DR: As Western automakers pull back from electric vehicles, Chinese rivals are accelerating — and history suggests this kind of hesitation rarely ends well. ⚡

Something slightly surreal is happening in the global car industry.

At the exact moment oil prices are rising (again), and EV demand is picking back up (again), Western carmakers are… easing off the accelerator.

Meanwhile, BYD and other Chinese players are doing the opposite — scaling fast, getting cheaper, and quietly taking market share across Europe and beyond.

If this feels familiar, it should.

Because the last time Western carmakers ignored a structural shift like this, Japanese manufacturers ate their lunch in the 1980s. History doesn’t repeat, but it does tend to rhyme. 🎶

The numbers tell the story.

• BYD has already overtaken Tesla as the world’s biggest EV seller
• European giants like Volkswagen and Stellantis are writing down tens of billions in EV investments
• Ford Motor Company has taken a $19.5bn hit and scrapped future EV models

In plain English:

👉 EVs are the future
👉 but right now, they’re less profitable
👉 so companies are retreating to protect short-term margins

Which makes perfect sense… if you’re thinking in quarters.

Slightly less so if you’re thinking about 2035. 😬

The irony is doing a lot of work here.

Western carmakers are saying demand is weak.

But demand is weak partly because:

• EVs are still expensive
• charging infrastructure is inconsistent
• and the best-value products are increasingly… Chinese

So instead of doubling down to fix that, many are hedging — investing in hybrids, keeping petrol alive, and waiting for the market to “settle.”

Spoiler: markets rarely settle in your favour while your competitors are scaling. 🧠

And then there’s policy.

The US has effectively slowed its EV push.
Europe is sending mixed signals.

So carmakers are stuck trying to invest in:
👉 electric
👉 hybrid
👉 and combustion

All at once.

Which, as one former exec put it, is basically “the worst of all worlds.”

Expensive, complex, and not particularly competitive.

Meanwhile, Chinese firms are doing something far less complicated:

👉 build EVs
👉 control batteries
👉 scale aggressively
👉 lower costs

No hedging. Just execution.

Which is why they’re not just winning at home — they’re expanding into Europe, India, Latin America… all the growth markets Western brands used to dominate. 🌍

So what?

For the auto industry: this isn’t just a technology shift — it’s a strategy test. And right now, Western players look like they’re optimising for survival, not leadership.

For investors and operators: short-term profitability vs long-term positioning is becoming the defining trade-off. The companies that hesitate may protect margins today… and lose relevance tomorrow.

For the UK and Europe: this is bigger than cars. It’s about industrial competitiveness. Lose the EV race, and you don’t just lose market share — you lose supply chains, jobs, and influence.

And for everyone else:

If your competitors are going all-in on the future…

…“waiting to see how things play out” is rarely the winning strategy. 🧩

Read more:
https://www.theguardian.com/business/2026/mar/21/west-carmakers-retreat-electric-vehicle-risks-irrelevance-iran-war-evs-china

VC is getting more diverse… slowly. But the real question is whether it’s enough. 💸

TL;DR: UK venture capital is becoming less male-dominated, but progress is real yet incremental — and the people holding power still look broadly the same.

There’s some genuinely good news here.

The share of UK VC firms with all-male investment teams has more than halved — dropping from 48% in 2017 to 21% in 2025.

And women now make up 31% of VC investment teams, up from just 18%.

Not vibes. Not pledges. Actual movement. 📈

And crucially, this didn’t happen by accident.

Frameworks like the Diversity VC Standard — now adopted by 100+ firms managing $44bn — are doing something the industry historically avoided: measuring itself.

👉 Firms using it saw 10%+ increases in gender diversity
👉 And even a doubling of Black representation at senior levels

That last bit matters.

Because diversity in entry roles is optics.
Diversity in decision-making roles is power. 🧠

There’s also a pipeline quietly doing real work.

Programmes like Future VC are placing 97% of participants into full-time roles across the ecosystem.

Which politely dismantles the industry’s favourite excuse:

“We just can’t find diverse talent.”

You can. You just weren’t looking properly. 🤔

Now, the reality check.

Even with all this progress:

👉 ~70% of investors are still not women
👉 senior leadership remains disproportionately white and male
👉 and capital still flows through familiar, often exclusive networks

And here’s the kicker:

Female founders still receive a tiny fraction of funding (single digits in many datasets).

So yes — teams are changing.

But outcomes? Much slower. ⚖️

So what?

For founders: access is improving, but who controls capital still shapes who wins. Representation upstream = opportunity downstream.

For funds: diversity is no longer a CSR line item — it’s a deal flow advantage. Different networks = different opportunities = differentiated returns.

For the ecosystem: this is what structural change actually looks like. Not viral moments. Not corporate pledges.

A decade of pressure, data, and forcing the industry to look in the mirror. 🔧

Closing thought:

The first decade made diversity visible.

The next decade decides whether it becomes unavoidable.

Read more:
https://sifted.eu/articles/diversity-vc-report-data-gender-ethnic-diversity

AI isn’t “coming for jobs.” It’s already reshaping them — and quietly removing the first rung of the ladder. 🪜🤖

TL;DR: 1 in 5 workers say AI is already replacing parts of their job, while ~16,000 jobs a month are disappearing in the U.S. — and it’s entry-level workers getting hit first and hardest.

We’ve officially moved past the hypothetical phase.

According to new data from Epoch AI and Ipsos, 20% of full-time workers say AI has already replaced parts of their job — not theory, not projections, but actual tasks disappearing in real time.

At the same time, Goldman Sachs estimates ~16,000 net jobs are being cut every month, as automation outpaces new roles created by AI.

So yes, AI is both:

👉 replacing work (20%)
👉 creating new work (15%)

…but the balance right now? Slightly negative. And very uneven. 📉

Because the real story isn’t “jobs vs no jobs.”

It’s which jobs are going first.

And surprise, surprise…

👉 admin
👉 customer service
👉 data-heavy junior roles
👉 anything vaguely repetitive but digital

In other words: entry-level work.

The exact roles people rely on to get their foot in the door. 🚪

This is where it gets uncomfortable.

We’re not just automating tasks…

We’re quietly removing the training ground for the workforce.

Those early roles used to do a few critical things:

👉 teach you how work actually works
👉 build experience (the thing every job requires but no job gives you)
👉 act as a bridge to higher-paying, more complex roles

If AI eats those… the ladder starts to wobble.

And the data is already pointing that way.

👉 Entry-level unemployment gaps are widening
👉 Wage gaps between junior and experienced workers are growing (~3.3%)
👉 Young workers are more exposed because they cluster in automatable roles

So while AI is making some workers more productive…

It’s also making it harder for others to even enter the game. 🎯

There’s also a slightly chaotic subplot here.

Half of workers using AI at work?

They’re doing it on personal accounts or free tools, not company-sanctioned ones.

So while execs are hosting “AI transformation strategy” offsites…

Employees are just quietly using ChatGPT in another tab and getting on with it. 😅

And then there’s the paradox.

Gen Z is:

👉 the most exposed to AI job loss
👉 and the most fluent in using AI tools

They’re being disrupted… but also adapting faster than anyone else.

Building side projects, using agents, learning on the fly.

Which means the long-term outcome might still favour them.

But in the short term? It’s messy. ⚖️

Zoom out and you see the real pattern:

This isn’t mass unemployment.

It’s a timing problem.

👉 AI destroys tasks quickly
👉 New roles emerge slowly
👉 Reskilling sits awkwardly in between

And right now, that gap is hitting the youngest workers hardest.

So what?

For individuals: the “graduate → entry-level → promotion” path is breaking. You’ll need to skip steps and prove value faster.

For companies: if you automate junior roles, you still need to answer a basic question — where do future seniors come from?

For policymakers: this is less about job loss, more about career pipeline collapse. Fixing that is much harder than just “creating jobs.”

Final thought:

AI isn’t replacing everyone.

It’s just starting with the people who haven’t had a chance yet.

Read more:
https://www.nbcnews.com/tech/tech-news/ai-job-work-replace-task-help-rcna267238
https://fortune.com/2026/04/06/ai-tech-displacement-effect-gen-z-16000-jobs-per-month/

📈 The tools behind the tech📉

📦Product📦

• Agentic Engine Optimization (AEO)
• Securing critical software for the AI era
  

📏Design📏 

• CodingFont: A game to help you pick a coding font 

👩🏿‍💻Code👩🏿‍💻

• Git commands I run before reading any code

🏢The business behind the tech🏢

• The unit economics of AI
• Are resumé's dead

📈 The tools behind the tech📉

Subject: Ready to celebrate LGBTQ+ Intersectionality with Just Eat Takeaway? 

Hi there,

Hope you’ve had a lovely start to the week and perhaps I can make it a little better by sharing our next event… 

On Thursday 23rd April, Just Eat Takeaway and Colorintech are coming together to celebrate the LGBTQ+ community. 

The event will feature a special fireside chat on how intersectionality is a superpower- leaving you feeling truly inspired. We'll also have workshops giving you a chance to connect with others in the room, alongside a Q&A, giveaways and more!

Did somebody say celebrity guest? 👀The panel will also feature an incredible LGBTQ+ champion within the music industry alongside incredible voices who have challenged their realities to come out the other side stronger. 

Check out the key details below: 

📅 Date: Thursday 23rd April 

🕕 Time: 15:30 - 18:30 (with time for networking) UK Time 

📍 Where: Central London

As this event has limited spaces, you’ll need to register your interest to attend using the form below: 

🔗 Application Link: https://luma.com/s85dsefo
🔴 Application Deadline: 20th April 2026

If you feel like this isn’t relevant for you but know someone who might love to join, feel free to forward this email to them ☺️

If you have any questions, please do not hesitate to reach out.

😃What we are consuming😃

🎉Claude mania

🤖Common AI terms

🗣️Google's new dictation app

🇫🇷France to ditch Windows for Linux

📟A game where you build a GPU