Nesan Pather

31/07/2025

Hey Facebook fam, feel free to check out some of my custom GPTs.

https://chatgpt.com/g/g-6887ac2d10948191925010d0f2e8dfbc-daily-habit-tracker

https://chatgpt.com/g/g-6887bcdd244c819183f1e3f334d478d4-easy-social

https://chatgpt.com/g/g-6888fa68a5c8819184302889b472406c-ai-news-pulse

https://chatgpt.com/g/g-688900b4b3288191886e6585553ea49c-chronicle

19/07/2025

18/07/2025

https://openai.com/index/introducing-chatgpt-agent/

ChatGPT now thinks and acts, proactively choosing from a toolbox of agentic skills to complete tasks for you using its own computer.

26/06/2025

🚀 AI’s Latest Leap: The Rise of Self-Learning Machines

The future of AI isn’t coming—it’s already here, and it’s teaching itself.

We’ve officially entered the age of self-learning AI—systems that evolve with little to no human input, capable of making sense of the world through experience and adaptation. Forget painstakingly labelled data. Think unsupervised learning, reinforcement rewards, and multimodal mastery.

Here’s what’s driving this shift:
🔹 Torque Clustering is helping AI find hidden patterns in raw, unlabelled data—perfect for decoding complex systems in medicine, finance, and science.
🔹 DeepSeek-GRM is bringing smarter, more human-aligned answers to the table through reward-based learning.
🔹 AI is now learning from sound + vision—watch out Hollywood, journalism, and creative industries.
🔹 Education, cybersecurity, and healthcare are already transforming with adaptive systems that learn and improve in real time.
🔹 And the next frontier? Agentic AI—digital assistants that don’t just react, but think, plan, and act on your behalf.

⚡️Why it matters:
✓ Faster innovation with less data
✓ Personalised, adaptive solutions
✓ Smart systems that get better with time
✓ AI that works with us, not just for us

We’re witnessing a seismic shift in what machines can do—autonomously. The only question is: are you ready to work with AIs that learn like humans?

09/06/2025

🚨 Apple Just Challenged a Core Belief About AI
A newly published paper from Apple’s Machine Learning Research team questions what many believe about AI’s ability to “think.”

🧠 The paper, titled “The Illusion of Thinking,” explores whether AI is truly capable of reasoning — or simply mimicking it.

For years, we’ve marvelled at AI’s ability to solve complex problems. But this research reveals a deeper truth: most large language models (LLMs) and even more advanced reasoning models (LRMs) might just be very sophisticated pattern matchers — not actual thinkers.

🔍 Key Takeaways:
• Pattern Matching ≠ Reasoning: AI models like Claude and DeepSeek perform well when tasks follow familiar patterns. But when complexity rises, their performance collapses.
• The “Complexity Wall”: Apple found that once problem difficulty crosses a certain threshold, even top models drop to near-zero accuracy.
• Three Complexity Levels:
• Low: LLMs sometimes outperform LRMs
• Medium: LRMs show an edge
• High: Both struggle or fail completely
• No General Problem-Solving: Models falter when faced with unfamiliar, logically demanding tasks — proving they don’t truly understand problems.
• More Scale ≠ More Intelligence: Simply making models bigger or prompting them better won’t unlock AGI. We need new architectures that prioritise real reasoning.

📌 Why This Matters
Apple’s study is a wake-up call. It reminds us to look beyond the hype and reassess what today’s AI is really doing. If we want AI to truly “think,” we’ll need to move past the illusion — and build systems that can handle complexity with true logic and adaptability.

🧠 A must-read for anyone working in AI, tech, or policy.

🔗https://ml-site.cdn-apple.com/papers/the-illusion-of-thinking.pdf

24/05/2025

Tech alone won’t align your GTM teams, but it can accelerate the right strategy.

Too often, I see organisations invest in powerful tools, but still struggle to scale.

Why?

Because the tools aren’t aligned with how people, processes, pipeline, and promotion connect.

At Huble, we design operational models that bridge strategy and systems.

Because alignment is the advantage.

24/05/2025

After a shaky start with Gemini. Google have firmly taken the mantle in the AI wars - some of these are just incredible

Learn more about the biggest announcements and launches from Google’s 2025 I/O developer conference.

23/04/2025

02/04/2025

No Hustle. Just…

Lifelong learning.
Systems.
Discipline.
Hard work.
Focus.
Over-deliver.
Be on time.
Be prepared.
Be relentless.
See the vision.

Hustle is loud.
Mastery is quiet.
We’re here for the long game.

24/03/2025

Waited some years to begin this project 🤗🙏🏼⚽️🥅

24/03/2025

Success in anything always hides in the details.

19/03/2025

Really excited for this event and I’ll definitely be there and looking forward to training with 👏🏼👌🥋