• Everyday AI
  • Posts
  • AI Market Forces Collide: Global Powers, Prices, and Problems (2025 AI Predictions - Vol. 2)

AI Market Forces Collide: Global Powers, Prices, and Problems (2025 AI Predictions - Vol. 2)

Anthropic unveils new agentic AI, Trump revokes AI safety order, U.K. unveils new chatbot and digital wallet, Claude’s new two-way voice mode and more!

Author

Everyday AI
January 21, 2025

Outsmart The Future

Sup y’all! 👋

Two down, three to go. 

In case you missed it, we’ve put together a literal can’t-miss guide to AI in 2025 in our five-part AI predictions and roadmap series. 

Here’s the lineup for the rest of the week: 

Vol 3. — Wednesday, January 22: Content's AI Revolution: Videos, Streams and Stars (Vol 3) 

Vol 4. — Thursday, January 23: Your Complete AI Toolkit: Tools, Rules, and Global Changes (Vol 4) 

 

Vol 5 — Friday, January 24: AI's Technical Leaps: Memory, Models, and Major Changes (Vol 5) 

Are you tuning in?

Login or Subscribe to participate in polls.

✌️
Jordan

Today in Everyday AI
6 minute read

🎙 Daily Podcast Episode: The AI market is going to be WILD in 2025. Major balance shifts are about to happen and people will be left behind. So here’s our predictions for the AI market this year. Give it a listen.

🕵️‍♂️ Fresh Finds: OpenAI’s product chief talks AI agents, NVIDIA and Mayo Clinic partnership and Mistral CEO speaks on AI in 2025. Read on for Fresh Finds.

🗞 Byte Sized Daily AI News: Anthropic unveils Virtual Collaborator agentic AI, Trump revokes AI safety order, U.K. unveils new chatbot and digital wallet as default and Claude’s new two-way voice mode. For that and more, read on for Byte Sized News.

🚀 AI In 5: We’re reviewing a data visualization feature from Poe and seeing how it stacks up against Claude Artifacts. See it here

🧠 Learn & Leveraging AI: Which AI companies will get left behind? How will the LLM battle turn out? We break down our predictions for the AI market in 2025. Keep reading for that!

↩️ Don’t miss out: Did you miss our last newsletter? We talked about AI super agents on the way, OpenAI’s Operator agent to release soon, Altman teases o3 mini and DeepSeek launches R1. Check it here!

 AI Market Forces Collide: Global Powers, Prices, and Problems (2025 AI Predictions - Vol. 2) ⚔️

The entire AI market is about to go super-friggin-nova in 2025.
 
↳  I'm not just talking about open source throwing hands with Big Tech (though THAT'S happening).

↳  I'm talking about tectonic shifts that are going to reshape who holds the power, who pays what, how LLMs will work and who's about to get LEFT BEHIND.

↳  From API prices falling through the floor to embodied AI stepping out of sci-fi and into your office – the next wave makes ChatGPT's launch look kinda like a kiddy splash pool.
 
I'm sharing the REAL market forces at play in Vol 2 of our 2025 AI Predictions series: AI Market Forces Collide.

Join the conversation and ask Jordan questions on AI here.

Also on the pod today:

• Prediction of Open Source Models 🔮
• Chinese AI Influence 🇨🇳
• Rise of Embodied AI 📈

It’ll be worth your 30 minutes:

Listen on our site:

Click to listen

Subscribe and listen on your favorite podcast platform

Listen on:


Upcoming Everyday AI Livestreams

Wednesday, January 22nd at 7:30 am CST ⬇️

Here’s our favorite AI finds from across the web:

New AI Tool Spotlight – Needle enables AI search across all your data, Rapport Studio lets you animate AI chatbots with your branding and Firecrawl helps you get structured web data.

OpenAI – OpenAI’s product chief says that the world is on the verge of AI agents

NVIDIA – NVIDIA and Mayo Clinic have partnered for AI-powered digital pathology.

Mistral - Mistral’s CEO spoke on how he believes the world will move past AI models and toward systems this year.

Google - Google DeepMind is eyeing clinical trials for designed drugs in 2025.

Cohere - Cohere has released an article on how its AI is trained for transparency and to provide citations.

AI in Society – Poland is upset over the current U.S. chip export and has expressed its frustrations.

AI Startups - Rekor Systems is working with SoundHound AI to revolutionize emergency vehicles with audio-visual AI.

AI in Sports – X Games is collaborating with Google to experiment with an AI judge.

1. Trump Revokes AI Safety Order 🇺🇸

President Donald Trump has revoked a key executive order from 2023 that aimed to mitigate the risks associated with artificial intelligence, originally set in motion by Joe Biden. This order required AI developers to share safety test results with the government, a precautionary measure aimed at protecting national security and public safety.

The Republican Party's platform cited the need for unimpeded AI innovation, suggesting that such regulations hinder progress and economic growth. As the landscape of AI continues to evolve rapidly, this shift raises concerns over potential impacts on job security and ethical standards in technology development

2. Anthropic Unveils New Agentic AI 🤯

In a recent reveal at Davos, Anthropic's CEO disclosed plans for a groundbreaking "Virtual Collaborator," designed to operate seamlessly alongside human employees by 2025. This AI will do everything from writing and compiling code to managing Slack conversations and updating Google Docs, essentially transforming how teams work together.

Anthropic is gearing up for a massive scaling effort, demonstrating confidence in overcoming the challenges that have historically hindered AI progress.

3. U.K. Government Unveils Digital Wallet and AI Chatbot 🇬🇧

The U.K. government has announced plans to launch a secure digital wallet for managing government-issued credentials, set to debut later this year. Alongside this, a new chatbot named ‘chat-uk,’ built in collaboration with OpenAI, will enhance user interactions on the GOV.UK portal, currently in testing, while a multi-functional AI assistant for government employees, dubbed "Humphrey," is also in development.

With the introduction of features like a virtual driver's license and enhanced age verification processes, these initiatives aim to streamline daily interactions while ensuring security.

4. Mistral AI Eyes IPO 👀

Mistral AI, a French startup founded by ex-Google DeepMind and Meta researchers, is gearing up for an initial public offering, according to CEO Arthur Mensch during an interview at the World Economic Forum in Davos.

With a hefty $1.14 billion raised and a valuation of around $6 billion, Mistral is positioning itself as a formidable competitor to U.S. giants like OpenAI, especially with its ChatGPT-like platform, Le Chat. The company also plans to expand its reach in the Asia-Pacific region with a new office in Singapore, hinting at ambitious growth strategies.

5. SandboxAQ Partners with Google Cloud for Large Quantitative Models 🚀

SandboxAQ has teamed up with Google Cloud to enhance its Large Quantitative Models (LQMs) and streamline their deployment via the Google Cloud Marketplace. This collaboration aims to accelerate advancements in critical fields like drug discovery and cybersecurity, potentially paving the way for breakthroughs in Alzheimer's research.

By leveraging Google’s top-tier cloud infrastructure, SandboxAQ is set to deliver transformative AI solutions more swiftly to businesses and governments alike.

6. Anthropic Unveils Two-Way Voice Mode for Claude 🗣

Anthropic's CEO Dario Amodei announced at the World Economic Forum that the company plans to roll out a "two-way" voice mode for its chatbot, Claude, along with a memory feature to enhance user interactions. This comes as Anthropic grapples with a surge in demand for AI and anticipates the launch of more advanced models in the near future.

Despite raising $13.7 billion, the company experienced significant losses last year and is in discussions to secure an additional $2 billion funding at a hefty $60 billion valuation.

Poe Previews - Better than Claude Artifacts?

There’s a feature from Poe called Previews and it’s pretty impressive!

We’re showing you how it works, when to use it and how it compares to Claude’s Artifacts feature.

🦾How You Can Leverage:

1 – Open Source AI Takes the Crown 👑

What that means:

Why it'll happen:

  • Meta's Llama models proved open source can compete with the big dogs. And NVIDIA’s Nemotrons doubles down on that point (check out our review on Lama 3.2 here).

  • The "Triple C" effect: Compounding Crowdsourcing of Compute is unstoppable

  • Companies are tired of sending their data to the cloud. Once Open Source catches up to proprietary this year in terms of benchmarks/human preferences, there will be ample reasons for companies to either switch to Open Source on the backend, or at least make it a plan B. 

2 – Chinese AI Dominates (And Disrupts) 🇨🇳

What that means:

  • 5 of top 10 global AI models will be Chinese-made

  • Alibaba's Qwen already has 90k enterprise customers and DeepSeek is getting sneaky good. (Emphasis on sneaky. Could be some Trojan Horse vibes here.) 

  • China has filed 6X more GenAI patents than the U.S. It’s only a matter of time until they are co-leaders in LLMs, robotics, video, AND audio

Why it'll happen:

  • Their humanoid robot production is already at warehouse scale. And pretty scary. 

  • They're crushing it with dirt-cheap API prices (see: DeepSeek. Also see: their terms & conditions. Yikes.)

  • The "Trojan horse" strategy: Offer irresistibly cheap and fast AI, collect WAY more valuable data

3 – Perplexity’s Big Decision 🔄

What that means:

  • Core ‘answers engine’ functionality is actually getting worse

  • Only 450k monthly searches vs ChatGPT's 30M. It’s really only the VCs who have poured in countless millions that are pumping Perplexity still 

  • They're scattered: daily podcasts, hardware plans, unfocused growth, “attempt” to merge with TikTok

Why it'll happen:

  • Google Deep Research puts them to shame (1,000+ sources vs 20)

  • ChatGPT and Claude keep improving their core products first

  • The answer engine space is suddenly PACKED. Perplexity will get squashed if they don’t pivot. 

We had Perplexity go head to head with its competitors in an AI search battle.

4 – API Prices Go Cliff Diving 📉

What that means:

  • GPT-4 already dropped from $60 to $2.50 per million tokens

  • Leading models will break the $0.50/million barrier

  • Edge AI adoption will accelerate the price drop

Why it'll happen:

  • Open source alternatives are forcing the market's hand

  • AI chips are getting more powerful AND more plentiful and cheaper. Like… you saw NVIDIA’s Digits my lawwwwwd

  • Cloud resources are being optimized like crazy

5 – Embodied AI Explosion 🤖

What that means:

Why it'll happen:

  • Pre-built AI makes hardware integration way easier

  • Regulations are finally catching up with the tech

  • Everyone except Tesla is actually delivering (maybe Tesla engineers are instead working on that new Grok that was supposed to come out in 2024?) 

We sat down with NVIDIA last year to discuss the future of robotics.

⌚

Numbers to watch

$125 Million

AI vision startup Metropolis acquires Oosto (formerly known as AnyVision) for $125M.

Now This …

Let us know your thoughts!

Vote to see live results

Do you attend our livestreams?

Every weekday, we bring you fresh AI insights, exclusive interviews, and breaking news with our Everyday AI livestream.

Login or Subscribe to participate in polls.

Reply

or to participate.