🎬DEI Is Dead. Long Live… Whatever Comes Next? 🪦

TL;DR: A former Goldman CEO just called DEI “counterproductive”—but the real story is how power is quietly rewriting the rules of inclusion. 🧩

Former Goldman Sachs CEO Lloyd Blankfein has labelled DEI programmes “futile” and “self-defeating,” arguing they “brand” participants as remedial. 🏷️

His comments land amid a broader rollback: major corporates—from Walmart to the banks at large—are scaling back DEI efforts following legal pressure and political shifts in the US. 🇺🇸

Blankfein’s critique taps into a growing narrative: that DEI, as implemented post-2020, created signalling without structural change. 📢

His core argument—that targeted programmes stigmatise participants—isn’t new. But it’s now being weaponised in a very different context: a post–affirmative action, anti-“woke” political climate where companies are retreating fast. ⚖️

Since 2023, legal rulings and policy shifts have made explicit diversity targets riskier. The result? Companies are rebranding inclusion as “skills-based hiring,” “merit,” or “broad talent development.” 🎭

But here’s the tension: DEI didn’t emerge in a vacuum. It was a response to systemic exclusion baked into hiring, promotion, and access. Remove the interventions without fixing the system, and you risk reverting to the baseline. 🔄

Meanwhile, not everyone is retreating. Apple, Costco, Delta, and Cisco are holding the line—suggesting this isn’t consensus, it’s a split in corporate strategy. 🧠

The uncomfortable truth: DEI is being reframed not because inequality is solved—but because the incentives around talking about it have changed. 💼

For talent: expect fewer explicitly labelled DEI programmes—but not necessarily more equitable outcomes. The work is going underground, not disappearing. 🕳️

For companies: this is a shift from visible commitments to invisible mechanisms. The risk? Less accountability, more ambiguity. 📉

And for readers building careers: don’t confuse the death of the label with the death of the problem. Access, networks, and bias still compound—just more quietly. 🔍

🔗 Read more

• Full story: https://fortune.com/2026/03/10/former-goldman-sachs-ceo-lloyd-blankfein-diversity-equity-inclusion-initiatives-donald-trump/

🔋Love in the Time of Algorithms (and Eventbrite) 💔🔋

TL;DR: Dating apps are splitting into two futures: AI that knows you better than your therapist, and IRL events that get you off the app entirely. 🤖

What happened

Bumble just unveiled “Bee,” an AI dating assistant that learns your values, goals, and communication style to act as a personal matchmaker. It’s moving beyond swiping into curated, AI-driven matches—and even experimenting with killing the swipe altogether. 🐝

Meanwhile, Tinder is going well two ways, 

On one hand it is: launching IRL events, virtual speed dating, and social-first features to get users meeting in real life again and on the other hand its launching a feature whereby its piloting a video speed dating experience in LA, where users can join scheduled three-minute video chats with potential matches. Sounds like a Chatroulette-ification of dating (yes, that still exists—proceed with caution). 🎥 I suppose some things just come back around into fashion

 Also: more AI under the hood to improve matching and safety. 🎟️

Let’s be clear: this isn’t innovation for fun—it’s survival. The dating app model is stalling, especially with Gen Z. 🧊

Bumble is betting on data depth over scale. Bee learns from private chats to build a richer user profile—meaning better matches, but also more intimate data capture than ever before. That’s not just matchmaking; that’s behavioural modelling. 🧠

Tinder, backed by Match Group’s $50M product push, is hedging differently: less screen time, more real-world interaction. Events, speed dating, and “vibe checks” are all attempts to fix what swiping broke—authentic connection. 🕺

Both are quietly powered by AI. Tinder’s “Learning Mode” and Bumble’s Bee aim to reduce friction and guess what you want before you even articulate it. That’s powerful—and slightly unsettling. 👀

The subtext? Dating apps are becoming social infrastructure platforms, blending AI, community, and commerce. And whoever wins owns not just dating—but how a generation forms relationships. 🏗️

So what🧠?

For users: expect fewer mindless swipes, more algorithmically curated relationships—and a trade-off between convenience and privacy. 🔐 It'll be a huge case study on if love can be scientifically and algorithmically curated

Read more

• Bumble’s AI “Bee”: https://techcrunch.com/2026/03/12/bumble-introduces-an-ai-dating-assistant-bee/
• Tinder’s IRL pivot: https://techcrunch.com/2026/03/12/tinder-tries-to-lure-people-back-to-online-dating-with-irl-events-virtual-speed-dating/

😪🍏 Meta’s AI Playbook: Fire the Humans, Hire the Agents 🧠😪

TL;DR: Meta is buying AI social networks, spending billions on compute—and maybe cutting thousands of jobs to fund it. 🚨

Meta has acquired Moltbook, a bizarre-but-viral “social network” where AI agents talk to each other, not humans. The founders are joining Meta’s Superintelligence Labs. 🤖

At the same time, Meta is reportedly considering cutting up to 20% of its workforce while ramping AI spend to eye-watering levels—up to $135bn in 2026, including a $27bn infrastructure deal with Nebius. 💸

Moltbook might look like a gimmick—but Meta’s interest is strategic. The key idea? An “always-on directory” of AI agents that can interact, coordinate, and potentially act on behalf of users. 🧩

Translation: Meta isn’t just building AI tools—it’s building AI ecosystems. Networks where agents don’t just assist you, but exist, communicate, and transact independently. 🌐 Just imagine a world where your IG ads are bought and optimised by an agent running on your instruction (Programmatic to the extreme); thats probably a dream Zuck has had too😅

Now layer that with the economics. Meta is doubling down on compute (data centres, chips, cloud deals) while reportedly eyeing massive layoffs. Analysts estimate a 20% cut could save ~$6bn—small change compared to its AI spend, but meaningful for margins. 📉

This is the real shift: labour → infrastructure. Humans are expensive, unpredictable, slower and pretty much the commodity in this. GPUs are not. 🖥️

Across the sector, 61,000+ job cuts have already been linked to AI. Whether AI is truly replacing roles—or just providing cover for cost-cutting—is still debated. But the direction of travel is clear. 🧭

Meanwhile, Meta is still behind OpenAI, Google, Anthropic on frontier models. So this isn’t dominance—it’s a high-stakes catch-up play. 🎯

So what🧠?

For talent: the bar is rising fast. Routine roles are under pressure; AI-adjacent skills are becoming table stakes. 📊

For builders: the opportunity isn’t just apps—it’s agent infrastructure, coordination layers, and new social primitives. Moltbook is weird, but it’s pointing somewhere real. 🧪

For everyone else: we’re entering a world where platforms don’t just connect people—they connect machines acting on behalf of people. That changes everything from work to identity to trust. 🔐

And yes—there’s an uncomfortable question: if AI agents become the primary users… are humans still the product? 🎭

Read more here

• Moltbook acquisition: https://arstechnica.com/ai/2026/03/meta-acquires-moltbook-the-ai-agent-social-network
• Layoffs & AI push: https://www.storyboard18.com/amp/digital/meta-shares-rise-on-report-of-potential-20-layoffs-amid-massive-ai-push-92387.htm
• $27bn infrastructure deal: https://www.bloomberg.com/news/articles/2026-03-16/meta-to-spend-up-to-27-billion-on-ai-infrastructure-from-nebius

🤖AI Isn’t Taking Your Job (Yet). It’s Taking Your Tasks. 🤖

TL;DR: AI can theoretically automate most white-collar work—but in reality, it’s only just getting started. 📊

Anthropic has released new research comparing what AI could do vs what it’s actually doing in the labour market—introducing a new metric: “observed exposure.” 🧪

The headline? While many jobs are highly automatable in theory, real-world adoption is still catching up. 🚧

The gap between hype and reality is doing a lot of work here. 🧊

Anthropic finds that roles in computer science, finance, and admin have over 90% theoretical AI exposure—meaning most tasks could be automated. But actual usage? Much lower. 📉

For example:
• Computer & math: ~94% theoretical vs 36% observed
• Admin roles: ~90% theoretical vs 34% observed
• Sales: ~62% theoretical vs 27% observed

That gap is the story. AI isn’t replacing jobs overnight—it’s quietly eating into workflows. 🍽️

At the individual level, the most exposed roles are telling:
• Programmers (74.5%)
• Customer service reps (70.1%)
• Data entry (67%)

In other words: structured, repeatable, language-heavy work is first in line. 🧾

Meanwhile, jobs requiring physical work or human presence—construction, transport, care—remain largely untouched (for now). 🏗️

Here’s the twist: the most exposed workers are more educated, better paid, and more likely to be women. Not the usual “automation hits low-skill jobs first” narrative. 🎯

And crucially—there’s no clear spike in unemployment yet. Instead, hiring for younger workers in these roles is slowing. Translation: disruption is happening at the entry point, not the exit. 🚪

So what🧠?

For early-career talent: the ladder is shifting. Roles that used to be stepping stones (admin, junior analyst, support) are being compressed or removed entirely. 🪜

For companies: this is less about layoffs and more about productivity arbitrage—doing more with fewer people over time. 📈

For everyone else: stop asking “Will AI take my job?” and start asking “Which parts of my job are already gone?” 🔍

Because the real shift isn’t job loss—it’s job redesign happening in slow motion. And if you’re not adapting, you’re already behind. ⏳

Read more

• Full report breakdown: https://www.euronews.com/business/2026/03/14/how-ai-will-reshape-work-anthropic-identifies-the-most-exposed-jobs

📈 The tools behind the tech📉

📦Product📦

• Levels of agentic engineering
  

📏Design📏 

• Turn Your Handwriting Into a Real Font — Free

👩🏿‍💻Code👩🏿‍💻

• The Future Of Software Engineering with Anthropic

🏢The business behind the tech🏢

• What do AI based layoffs say about their ability to scale?
• Learnings from Paying Artists Royalties for AI-Generated Art

🙌Colorintech // Canva🙌

Join us on Wednesday, 25th March, at 12:30pm GMT for our IWD event, a virtual event tailored to everyone interested in gender equity, data visualisation and cutting cutting-edge technology.

Hosted in collaboration with Colorintech and Flourish, this one-hour session will provide you with:

• An exclusive look into the groundbreaking stats on Gender equity
• Insights into the tools flourish use to bring data to life.
• An interactive q&A discussion featuring distinguished Tech professionals

Register here

🙌A data event🙌

Kickstart your career in the AI Data Cloud! Join BDPN for a free, instructor-led webinar designed for beginners. Gain hands-on experience with Snowflake architecture, data integration, and Cortex LLM functions.

• When: Friday, 20th March | 9 AM – 1 PM (Online)

• Cost: FREE

• Perk: Complete the session to receive a Snowflake Platform Associate certification voucher (worth $100).

What You’ll Learn:

• Navigating the Snowflake UI and Notebooks.

• Building database and compute resources.

• Leveraging Generative AI via Snowflake Cortex.

• Securely loading and sharing data.

Who should attend? Anyone with basic database knowledge and limited Snowflake experience, from students to aspiring Data Engineers and Scientists.

👉 Register Here to Secure Your Spot

Questions? Reach out to Ian Benjamin at info@theBDPN.com.

😃What we are consuming😃

👀OpenAI to Cut Back on Side Projects in Push to ‘Nail’ Core Business

🎧Spotify will let you edit your Taste Profile to control your recommendations 

📚Encyclopedia Britannica sues OpenAI over AI training 

🤖Robots on the pavement?