Old Dogs, AI Tricks: Why Legacy Systems May Win the AI Race

TGIF!

Here’s a spicy thought to take into Friday: What if the most “AI-ready” companies aren’t the ones launching internal GenAI labs or hiring Prompt Engineers 2.0, but the ones still running on Oracle databases and collecting 30 years of seemingly boring ops data?

Yep. The same companies we used to tease for their legacy systems are now quietly building AI moats the size of continents. In today’s edition, we dive into why that matters, how China’s AI heavyweights are sabotaging each other, and what Sundar Pichai has to say about surviving the singularity.

Let’s get into it.

Start learning AI in 2025

Keeping up with AI is hard – we get it!

That’s why over 1M professionals read Superhuman AI to stay ahead.

• Get daily AI news, tools, and tutorials

• Learn new AI skills you can use at work in 3 mins a day

• Become 10X more productive

Sign up and start mastering AI

How Legacy Companies Are Quietly Winning the AI Game

While everyone’s been chasing the next GenAI unicorn, the true AI goldmine might just be. your grandpa’s ERP system. Legacy companies, the ones often mocked for their 90s tech and three-year upgrade cycles, are now leapfrogging their startup competition thanks to one thing: data gravity.

A new report from Fast Company showcases how organizations like Caterpillar, 3M, and JPMorgan are quietly turning their massive, decades-old data stores into competitive weapons. Forget shiny new AI toys, these giants are training LLMs on proprietary, structured, high-fidelity data that no one else has. And it’s changing the game.

Why enterprises should care:

• Your past is your power. All that historical transaction data is now a training set.

• Proprietary data = differentiated AI. Public models are great, but they’re generalists. Your enterprise data lets you fine-tune for real edge.

• Legacy ≠ laggard. If you’ve got good governance and structured workflows, you’re probably more AI-ready than you think.

As AI matures past the experimentation phase, the orgs that own the best training data, not just the best tools, will be the ones to beat.

What to Watch: China’s AI 'Dragons' Are Cannibalizing Each Other

China’s biggest AI players (Baidu, Alibaba, Tencent) are all sprinting toward the same goal: dominance in foundation models. But are they racing into a chokehold?

The problem:

• Too many models, too little differentiation.

• Internal competition is stifling ecosystem collaboration.

• The Chinese government’s “Big Model Innovation Alliance” risks becoming more bureaucracy than breakthrough.

Why this matters for global enterprise strategy:

• China's AI turmoil = opportunity for Western vendors to differentiate with vertical AI products.

• Enterprises with international supply chains need to watch China’s internal dynamics. They could impact tooling and availability.

• It’s a reminder that not every AI arms race ends in victory. Sometimes it ends in gridlock.

In the News

1. Deepfake Dominoes: AI Voice Scam Hits Bank in UAE
A BBC report reveals a shocking case where a bank manager was duped by a deepfake voice into authorizing a $35 million transfer. Spoiler: it worked.
→ Full story

2. YouTube’s AI Search Just Got Weirder (and Smarter?)
YouTube is testing AI-generated summaries for comments and video content. Expect weirder, faster recommendations... and probably a few existential spirals.
→ Get the update

3. Google CEO: AI Doom Is ‘Pretty High’ But Let’s Hope For the Best
In a moment of optimistic nihilism, Sundar Pichai said the chances of AI extinction are “pretty high”—but humanity might rally. Confidence-inspiring? Not really.
→ Read more

TL;DR:

• Legacy companies are AI powerhouses thanks to rich proprietary data. Don't sleep on them.

• China’s AI giants are in a turf war, risking innovation gridlock.

• AI news you need: Deepfake fraud hits banks, YouTube goes AI-native, and Google’s CEO ponders the apocalypse.

Stay sharp,

Cat Valverde
Founder, Enterprise AI Solutions
Navigating Tomorrow's Tech Landscape Together