In partnership with

Good Morning! This week in tech, a court battle over a glitchy gambling payout ended with Paddy Power being forced to hand over £1M to a lucky (and very patient) winner. Meanwhile, China’s hyped AI assistant Manus promised big things but mostly delivered broken links and frustration. And in a major win for computer science, a Rutgers undergrad just smashed a 40-year-old data structure conjecture, proving that hash tables can be way faster than we ever thought.

— Forrest Knight & Meghanadh Vasireddy

Paddy Power’s £1M Gamble Gone Wrong

Context: A recent High Court ruling has left betting giant Paddy Power on the losing side of a £1M dispute. In 2020, Corrine Durber, a gardener from Gloucestershire, hit the “Monster Jackpot” while playing Wild Hatter online. However, the company later claimed she had only won £20,265 due to a software error. Naturally, Durber wasn’t buying that explanation and took the fight to court.

What’s New: This week, Justice Ritchie ruled in her favor, emphasizing that “what you see is what you get” should be the gold standard in digital gaming. Paddy Power had argued that the real outcome was determined by a random number generator (RNG) and that a display bug misrepresented the winnings. But the court wasn’t having it—the game showed £1M, and that’s what she should get.

Takeaways:

• Accountability in Software – Bugs in financial systems, whether in banking or gambling, aren’t just inconveniences—they impact real people’s lives.

• Legal Precedent for Digital Contracts – The ruling reinforces that UI/UX errors don’t excuse companies from honoring payouts.

• Better Testing Needed – Paddy Power admitted that similar display issues had affected other players. That’s a major QA failure.

This case is a wake-up call for online gaming platforms—when real money is on the line, transparency is non-negotiable.

Optimize global IT operations with our World at Work Guide

Explore this ready-to-go guide to support your IT operations in 130+ countries. Discover how:

• Standardizing global IT operations enhances efficiency and reduces overhead

• Ensuring compliance with local IT legislation to safeguard your operations

• Integrating Deel IT with EOR, global payroll, and contractor management optimizes your tech stack

Leverage Deel IT to manage your global operations with ease.

Download free guide

Manus: AI Hype or Just Another Overpromised Assistant?

Context: The AI world has been buzzing about Manus, a new “agentic” AI platform claiming to revolutionize automation. Developed by Chinese startup Monica, Manus allegedly performs high-level tasks like buying real estate, executing financial trades, and even programming video games. But is it really the next big thing, or just another case of AI overpromising and underdelivering?

What’s New: Manus has attracted massive hype, with AI influencers calling it “the most sophisticated AI tool” and Discord invites selling for thousands of dollars. But early users report a different story:

• Error-prone execution – Simple tasks like booking flights and ordering food often result in crashes or incomplete actions.

• Limited true autonomy – While Manus integrates multiple AI models (Claude, Qwen), it doesn’t appear to be a fundamentally new architecture.

• Marketing vs. reality – Viral videos showcasing its capabilities were later debunked by its own developers.

Like many AI products, Manus is promising the world but delivering a handful of broken links. It’s still in early access, so improvements may come—but for now, it's more Deep Disappointment than DeepSeek Moment.

Breaking a 40-Year-Old Computer Science Conjecture

Context: Imagine casually reading an old research paper and accidentally rewriting a fundamental rule of computer science. That’s exactly what Andrew Krapivin, then an undergraduate at Rutgers, did. While exploring a concept called Tiny Pointers, he stumbled onto a new way to structure hash tables—one of the most studied and widely used data structures in computing.

But here’s the kicker: his discovery shattered a 40-year-old conjecture by legendary computer scientist Andrew Yao, which had defined the theoretical speed limits of hash table operations since 1985.

What’s New: Krapivin, now a grad student at Cambridge, worked with researchers Martín Farach-Colton (NYU) and William Kuszmaul (CMU) to prove that their new hash table design finds elements exponentially faster than Yao’s model predicted.

• Faster Lookups – The worst-case search time improves from O(x) to (log x)², a game-changing boost.

• Constant-Time Searches – Under certain conditions, queries can be done in constant time, completely independent of table fullness.

• Rethinking Fundamental Limits – This doesn’t just disprove Yao’s conjecture—it provides a new best possible bound for a core data structure.

While practical applications are still unclear, this breakthrough proves we still have a lot to learn about optimizing fundamental data structures.

🔥 More Notes

• Seagate's Breakthrough in Data Storage Technology: Seagate has unveiled a significant advancement in data storage with its heat-assisted magnetic recording (HAMR) technology. This innovation uses a laser smaller than a grain of salt to encode more data onto hard disk drives, enabling storage capacities of up to 36 terabytes. As artificial intelligence applications drive increased data storage demands, Seagate anticipates substantial growth in sales and earnings. While solid-state drives (SSDs) offer faster speeds, they remain more expensive per terabyte, ensuring that hard drives will continue to play a crucial role in data storage for the foreseeable future.

• Nanorobots: The Future of Targeted Cancer Treatment: Nanorobots, entities smaller than blood cells, are showing promise in revolutionizing cancer treatment by enabling precise, targeted therapies with fewer side effects. Unlike traditional chemotherapy, which affects both cancerous and healthy cells, nanorobots made of biological materials like RNA or DNA can be programmed to deliver drugs directly to tumor sites. Recent research at the Karolinska Institute demonstrated the effectiveness of nanorobot-based chemotherapy in mice, where DNA origami techniques released drugs specifically at tumor sites. While mainstream medical application of nanorobots is still years away, their potential to detect and treat conditions like cancer and liver disease early could transform healthcare, making concepts from science fiction conceivable realities.

📹 Youtube Spotlight

Was this forwarded to you? Sign Up Here