Alex Banks

Building a better future with AI

Alex Banks is a founder, creator and investor. He reaches over 1 million people on social media each week with insights into the world of artificial intelligence (AI).

He is the founder and writer of The Signal, a weekly newsletter that cuts through the AI noise to over 50,000+ readers.

Alex is also an active early-stage investor and advisor to startups across the AI and technology landscape.

👉 All partnerships, enquiries and speaking engagements: contact@noisemedia.ai

Writer

As an introvert, I used to struggle with interaction.

Luckily AI is becoming the Great Equaliser:

1. Speech-to-speech: Real-time conversation without the pressure
→ Example: ChatGPT's voice mode enables natural, flowing dialogue
→ Benefit: Perfect for on-the-go brainstorming, notetaking or setting tasks
→ Access: https://chatgpt.com/

2. Visual learning: Complex information made simple
→ Example: NotebookLM creates mind maps, flashcards, and video overviews
→ Benefit: Transforms dense material into formats that match how you think
→ Access: https://notebooklm.google/

3. Content creation: Building your personal brand becomes accessible
→ Example: GrowHub crafts LinkedIn posts that sound like you, not AI
→ Benefit: Share your thoughts online without hours of writing
→ Access: https://growhub.ai/

4. Speech-to-text: Your thoughts flow without typing friction
→ Example: Wispr Flow captures your ideas through natural dictation
→ Benefit: Write faster by speaking, perfect for those who think out loud
→ Access: https://lnkd.in/eCTTPPEz

5. Guided learning: Ask unlimited "why's" without judgment
→ Example: Gemini 2.5 Flash's guided learning mode breaks down any topic
→ Benefit: Patient, step-by-step explanations at your own pace
→ Access: https://gemini.google.com/

My take?

The possibilities of what we can achieve with AI are endless.

Especially when paired with human creativity, curiosity, and ambition.

A whole new game has been created where everyone gets to play.

So here's my question to you:

How has AI helped you overcome a personal challenge?

Share your story below.

P.S. Follow me Alex Banks for daily AI highlights & insights.

I cover the most important AI developments each week in my newsletter.

Subscribe here: https://lnkd.in/ePSZP6KF

Image credit: Tobi Lutke on X

Google just woke up.

And OpenAI is feeling the heat.

A leaked internal memo from Sam Altman just surfaced.

The OpenAI CEO admits Google's AI progress could "create some temporary economic headwinds."

Here's what's happening:

→ OpenAI's revenue growth could slow to single digits by 2026
→ They're projecting a $74 billion operating loss by 2028
→ Gemini 3 Pro now leads GPT-5.1 in reasoning and coding tasks

Altman's own words: "Google has been doing excellent work recently in every aspect."

This is a massive shift.

For 3 years, OpenAI had peacetime dominance.

Now it’s a wartime scenario.

OpenAI is scrambling with social-infused features like Sora 2 App and group chats in ChatGPT in an effort to push top-line revenue and seek virality.

Meanwhile, Google's been quietly building:

→ Gemini 3 for reasoning
→ Nano Banana Pro for images
→ Veo 3.1 for video
→ TPUs powering everything

The real insight here:

• Google wasn't the first search engine
• Many came and went before it
• What matters is being the last mover

Owning the "stack" from chip to chatbot creates the only real enduring moat.

My takeaway:

Frontier LLMs are rapidly commoditising.

OpenAI pioneered the technology, but it can't defend margins when competitors match its capability.

I recently switched my new default model from GPT-5.1 to Gemini 3 “Fast” mode for general Q&A tasks because of its speed.

The sleeping giant has woken up.

And the AI race just got a whole lot more interesting.

Follow me Alex Banks for daily AI highlights and insights.

I cover the most important AI developments each week in my newsletter.

Subscribe here: https://lnkd.in/ePSZP6KF

Runway just dropped Gen-4.5.

It's now the world's # 1 AI video model.

The numbers:

→ 1,247 ELO score on Artificial Analysis leaderboard
→ Beats Google’s Veo 3, Kling 2.5, and OpenAI’s Sora 2 Pro
→ Top-rated model globally

Core capabilities:

→ Unprecedented physical accuracy (realistic weight, momentum, and force)
→ Complex multi-element scenes rendered with precision
→ Maintains visual consistency across diverse aesthetics
→ Intricate camera choreography from single prompts
→ Handles both photorealistic and stylised animation

What's different:

• Built entirely on NVIDIA GPUs (Hopper and Blackwell series)
• Same speed and pricing as Gen-4
• All existing control modes coming to Gen-4.5 (Image to Video, Keyframes, Video to Video)

Current limitations:

1. Causal reasoning issues: A door might open before someone touches the handle. Effects can precede causes.
2. Object permanence problems: A cup can vanish after being blocked from view, then reappear in the next frame.
3. Success bias: Actions succeed too often, e.g., a poorly aimed basketball shot still goes in.

Already being used by:

→ Target (Retail & Ecommerce)
→ BBC (Broadcast)
→ Ubisoft (Gaming)

Runway says it will be available to everyone in the coming days.

My takeaway:

Two years ago, Runway released Gen-1, the first publicly available video generation model.

Now they're setting a new benchmark.

We’re one step closer to systems that truly understand how the real world actually works.

I can feel the video generation race really accelerating and we're all going to benefit from this competition.

Follow me Alex Banks for daily AI highlights and insights.

Want to learn how to leverage AI in your work?

Subscribe to my newsletter:
https://lnkd.in/ePSZP6KF

China just shipped $113M of humanoid robots.

UBTECH’s Walker S2 has begun mass production:

The Numbers:

→ 800M yuan (~$113M) in orders secured
→ 500 robots delivering by year-end
→ 5,000 annual capacity by 2026
→ 10,000 units by 2027

Who's Buying:

• BYD
• Geely Auto
• FAW-Volkswagen
• Foxconn
• Dongfeng Liuzhou Motor
• SF Express

What Makes Walker S2 Production-Ready:

→ 176cm tall, 70kg, 52 degrees of freedom
→ Walks at 2m/s with humanoid gait
→ 162° waist rotation for tight spaces
→ Deep squat capability with 15kg payload
→ Gen-4 dexterous hands with 6 tactile sensors
→ "Human-eye" binocular vision system

I was impressed by the 3-minute autonomous battery swap:

Traditional robots:
Work 5 hours → Stop → Charge → Wait

Walker S2:
Work 5 hours → Swap battery (3 mins) → Keep working

The dual-battery system monitors power in real-time and autonomously chooses to swap or charge based on task priorities.

Result: 24/7 continuous operation with zero human intervention.

Market Response: 

→ UBTECH's stock climbed 150% this year to 133 HKD
→ Revenue up 27.5% to 621M yuan
→ Humanoids now represent 30% of total sales (up from 10% last year)

My takeaway:

We've crossed the chasm from R&D to real deployment.

The race has shifted from who can build the flashiest demo to who can deliver real scale, maintain uptime, and prove ROI.

UBTECH seems to be answering all three.

Follow me Alex Banks for daily AI highlights and insights.

P.S. I’ll be doing a full breakdown in my newsletter this week.

I cover the most important AI developments to keep you ahead.

Subscribe here: https://lnkd.in/ePSZP6KF

This 1965 clip of Isaac Asimov blows my mind.

He predicted exactly what we're building today:

The convergence of humans and machines into something new.

The two-way transformation is now undeniable:

1. Robots becoming more human-like

• Humanoid robots are starting to move like us
• Neural networks are mimicking our brains
• Soft robotics using organic-inspired materials

2. Humans incorporating more technology

• Neuralink developing brain-computer interfaces
• Advanced prosthetics controlled by thought
• Artificial organs replacing failing human ones

Most people only know Asimov for his famous Three Laws of Robotics:

1. A robot may not harm a human or allow harm
2. A robot must obey human orders (unless violating Law 1)
3. A robot must protect itself (unless violating Laws 1 or 2)

Asimov saw what others couldn't.

While everyone debated if machines could think, he asked what happens when they merge with us.

A hybrid species might surpass both humans and machines.

My takeaway:

In a world obsessed with AI replacing humans, Asimov offered a different vision 60 years ago.

Not replacement, but convergence.

The line between humans and machines will continue to blur.

Is this the future we need?

Follow me Alex Banks for daily AI highlights and insights.

I cover the most important AI developments each week in my newsletter.

Subscribe here: https://lnkd.in/ePSZP6KF

The Fairfield Company

Founder

Colonial Pipeline Company

Corporate Development

Fortress Investment Group

Analyst - Liquid Markets

UBS Wealth Management

Summer Analyst

Founder of an alternative investment firm focused on assets valued between $2 & $10 million.

taplio

The Battle

Alex Banks

Chris Munn

Followers

Posts

Engagements

Likes

Performance ScoreAlex Banks's LinkedIn Performance

Ready to grow your LinkedIn presence?

taplio

The Battle

Alex Banks

Chris Munn

Followers

Posts

Engagements

Likes

Performance Score.css-goub6s{font-size:var(--chakra-fontSizes-md);font-weight:500;color:var(--chakra-colors-gray-600);display:block;margin-top:var(--chakra-space-1);}Alex Banks's LinkedIn Performance

Ready to grow your LinkedIn presence?

Performance ScoreAlex Banks's LinkedIn Performance