You’re receiving this email because you subscribed to the newsletter. If you’d like to unsubscribe, you can manage your preferences here with one click (or use the button at the bottom of the email).

Welcome to this edition of Loop!

To kick off your week, I’ve rounded-up the most important technology and AI updates that you should know about.

‏‏‎ ‎ HIGHLIGHTS ‏‏‎ ‎

Let's jump in!

1. Anthropic will pay authors $1.5 billion to settle a lawsuit over AI training data

We start with Anthropic this week, which has agreed to settle a lawsuit over AI training data and will pay authors at least $1.5 billion. If this settlement is approved by the courts, those authors will receive around $3,000 per book.

Sounds like a fantastic deal for Anthropic, who have just raised $13 billion from investors and are now valued at $183 billion. For authors, it’s a minuscule amount.

The saga began a year ago, when several authors accused Anthropic of building its "multibillion-dollar business by stealing hundreds of thousands of copyrighted books."

A federal judge initially ruled that training on legally purchased books constituted fair use, but Anthropic still faced allegations about using pirated materials.

By July, authors had won the right to bring a class action representing all US writers whose work allegedly came from pirated libraries.

Crucially, this settlement only covers past actions. It doesn't grant Anthropic any future training rights, and the company must destroy all downloaded files and copies.

2. Amazon’s AI can now find items with your camera

Called Lens Live, this new shopping feature can now identify items that you’re looking at. After you’ve pointed your camera at the item, the AI will try and search for it across billions of marketplace listings.

As of right now, the feature is exclusive to the Amazon Shopping app on iOS. Once it has found the product, you can easily add it to your cart and buy within seconds - essentially creating a purchase option for everything you see.

To achieve this, the team used a computer vision model to identify the item and then connected it with Amazon’s huge database of products. It also taps into Rufus, Amazon's AI assistant, which will offer a product summary and can answer questions about the item you’re looking at.

The timing feels particularly strategic as Amazon appears to be responding to Google's Gemini Live. Google’s product offers very similar features, but it has more of a focus on answering questions, rather than purchases.

While this is incredibly convenient and will allow us to get answers about a product in real-time, it will blur the line between browsing and buying even further - making impulse purchases all the more likely.

3. CoreWeave acquires a startup that trains AI agents

The company has acquired OpenPipe, a startup that helps enterprises to build custom AI agents through reinforcement learning.

If you haven’t heard of CoreWeave before, they are a huge cloud company that has really benefited from the AI boom - with OpenAI among its biggest clients.

OpenPipe had already made waves with ART (agent reinforcement trainer), an open-source toolkit that's gained traction amongst developers building AI agents.

Their approach centres on reinforcement learning, which rewards AI models for correct responses, and is a powerful way to fine-tune AI models. This ensures that the results are closer to your business needs, but it can be somewhat expensive to run.

This acquisition is part of CoreWeave's continued push to expand beyond pure infrastructure. It seems like they're in the process of building a more comprehensive platform for AI training, as they also snapped up an AI developer platform earlier this year.

CoreWeave will likely try to make the reinforcement learning process even easier, as they'll be providing both the raw compute power and the tools needed to use it effectively.

It's a solid acquisition in an industry where the lines between infrastructure and application layers are increasingly blurred.

4. Runway prioritises robotics and self-driving cars

The New York-based company has spent seven years perfecting its AI world models for filmmakers and artists, with Gen-4 and Runway Aleph becoming go-to tools for video generation and editing.

But robotics and autonomous vehicle companies have started to knock on Runway's door, essentially asking if they could borrow the company's reality-simulating tech for their own purposes.

This is something that I’ve regularly flagged in this newsletter, as training robots and self-driving cars in the real world is incredibly expensive, time-consuming, and difficult to scale.

In fact, Google’s Waymo does something similar. Their self-driving cars have sensors onboard, which are used to create 3D maps and digitally re-create the world around them.

This is incredibly valuable, as these simulations allow engineers to test algorithm updates and see how the vehicles perform in various weather conditions.

Rather than building entirely new products, Runway plans to fine-tune its existing models for these industries and will create a dedicated robotics team.

Their nearest competitor will be Nvidia’s Cosmos, but Runway is hoping that their filmmaking experience will give them an advantage in simulating the real-world.

5. Google reveals an AI tool for screenwriters

Google DeepMind has unveiled Fabula, which is a new AI writing assistant for screenwriters and playwrights.

Rather than churning out complete stories, the tool will act more like a creative partner and will suggest different ideas for the story - with writers in full control over the narrative vision.

Using their latest version of Gemini, the app helps writers to iterate through different versions of characters, plot points, and story arcs - with AI suggestions that actually make narrative sense.

DeepMind’s team has run extensive workshops and gathered feedback from those in the industry - including screenwriters and producers - on how the tool is being used. They’re now expanding that access to more writers.

If it sounds like something you or your team could benefit from, I’ve included a link below.

At Workday Rising next week?

I’ll be at the conference representing Kainos, as we reveal our first AI agents for Workday. My team has been working on these agents for several months, alongside other early-access partners.

We will be revealing three new AI agents in San Francisco. These agents will help you to organise travel bookings in minutes, instantly resolve help desk queries, and identify the AI features that will benefit your business.

You can visit our stand to experience hands-on demos, speak with our experts about your specific challenges, and learn how you can deploy AI agents - freeing your teams to focus on what matters most.

Or you can book a 30 min session with me, using the link below. Simply select “Workday Marketplace Apps” and fill in your details.

OpenAI is creating a hiring platform to take on LinkedIn

The company is stepping into LinkedIn's territory, with plans for an AI-powered hiring platform that will match workers with companies.

The OpenAI Jobs Platform, which is expected to launch in the middle of 2026, is part of the company's effort to diversify beyond ChatGPT.

Professionals can also certify their AI skills within the platform and add them to a job application. The certification programme already has Walmart on board, with OpenAI ambitiously targeting 10 million certified Americans by 2030.

But this move is unlikely to reduce tensions with Microsoft, who own LinkedIn and saw its revenues rise by 9% last year.

Of course, there is the real risk that AI - and OpenAI’s technology in particular - will be used to eliminate millions of white-collar jobs. But we don’t know if that will happen for sure, or when exactly that will be the case.

Anyone who says otherwise is either selling something, or not being honest about the challenges involved. For OpenAI, it seems like a solid move and will allow them to become the central place for both businesses and workers.

It will also allow them to have access to incredibly valuable data about the labour market and how AI could be used to automate specific jobs. This will help them to make strategic decisions on new products.

Microsoft already does this, as they use LinkedIn’s data to create trend reports on the workplace and can identify jobs that are in high demand.

This new platform could also open an extra revenue stream for OpenAI, which is in desperate need of cash to continue investing in AI models and infrastructure.

Just a few days ago, the startup announced that it expects to burn through $115 billion by 2029 - a truly staggering amount that could be partially funded with the OpenAI Jobs Platform.

🚗 Tesla proposes new pay package for Elon Musk, could be worth $1 trillion

🤖 This clever robot mower can also pick fruit and play fetch with your dog

💻 OpenAI snaps up the team behind Xcode coding tool Alex

🧠 DeepSeek plans to unveil its next AI model before 2025 ends

🇨🇭 Switzerland launches an open-source AI model

⚡ The staggering numbers behind China’s dominance of the EV industry (NYT)

✈️ GE Aerospace invests $300 million in Beta Technologies, will develop hybrid-electric aircraft

🌐 Atlassian acquires Arc browser company for $610 million

💔 Godfather of AI says his girlfriend used ChatGPT to break up with him (I’m not even joking)

🏛️ Google avoids a costly breakup with Chrome and Android

Surprisingly, most farms still rely on tiny crop samples and then use this data to make million-dollar decisions about their harvest.

Orchard Robotics is hoping to improve this, with their cameras mounted on the tractor and used to capture high-resolution images of the fields.

Its AI will then analyse these images for fruit size, colour, and other health indicators - providing the farmer with detailed data about their crops and how healthy they are.

The technology has already been deployed in the US, with the company initially focusing on large apple and grape harvests. They’re now hoping to expand further and allow the system to work with blueberries, cherries, almonds, and other crops.

The startup has just closed a $22 million Series A, which should help the team to grow faster and further improve the product.

For an industry where margins are incredibly tight and there are plenty of uncertainties with the weather, this data could give some farms a competitive advantage and optimise some of their operations.

This Week’s Art

We’ve covered quite a bit this week, including:

If you found something interesting in this week’s edition, please feel free to share this newsletter with your colleagues.

Or if you’re interested in chatting with me about the above, simply reply to this email and I’ll get back to you.

Have a good week!

Liam

Share with Others

If you found something interesting in this week’s edition, feel free to share this newsletter with your colleagues.

Liam McCormick is a Senior AI Engineer and works within Kainos' Innovation team. He identifies business value in emerging technologies, implements them, and then shares these insights with others.