Hey Admin ,

Welcome to this week’s new(s)line report. There's been more announcements, corporate moves, and – everyone's favourite – lawsuits this week on the AI scene. Let's go take a look shall we?

Today’s headlines:

Let's dive right in.

LATEST DEVELOPMENTS

Another day, another lawsuit - is Disney's Midjourney case any different?

“Don’t mess with the mouse.”

This phrase is common knowledge amongst media enthusiasts and lawyers alike. Even South Park made fun of it (warning for strong language…obviously).

Disney isn’t just a corporation with good IP lawyers, they’re a company that permanently altered copyright law to let them hold on to the rights for its mascot – Mickey Mouse.

And Disney had a new target in its sights.

Midjourney.

Disney and its co-plaintiffs allege that the AI startup has been plagarising it since its inception, using copyrighted characters in its image generation – and it seems like Disney have them dead-to-rights.

You can read the filing here, but to summarize:

• Disney refers to Midjourney as a ‘bottomless pit of plagiarism’ and claim it’s infringment is ‘calculated and wilful’.

• The case includes multiple examples of Disney characters showing up in the output of Midjourney – mostly from Star Wars like the image above, but also Simpsons, Marvel and Pixar characters.

• There are dozens of examples, all showing Midjourney output and Disney IP side by side – and they’re all very much undeniably the same characters.Other companies and subsidiaries are listed as co-claimants in the suit, for example Universal Studios, Dreamworks, Lucasfilm.

• The suit claims that Midjourney not only allows and promotes plagiarism, but that Midjourney has marketed its own service using plagiarised images.

It doesn’t help Midjourney’s case, or public perception, that a damning excerpt of an interview with CEO David Holz has resurfaced:

Forbes: Did you seek consent from living artists or work still under copyright?

Scholz: No. There isn’t really a way to get a hundred million images and know where they’re coming from. It would be cool if images had metadata embedded in them about the copyright owner or something.

That’s…uhhh…not a great look.

And this much evidence sitting in the lap of Dinsey’s mega-lawyers, possibly the single best IP protection force on the planet.

This is probably the first time an AI company has been so clearly “the little guy” in one of these lawsuits.

In other words, if Disney doesn’t win this case, the question of AI plagiarism is legally answered in the US - because if they don’t win, nobody else will.

OpenAI launches o3-pro and slashes price

OpenAI just dropped o3-pro –a high-performing version of its famous reasoning model that competes with the other top dogs– and reduced prices by 80%.

(This price reduction is not on the blog, it was announced by Sam Altman on X.)

This model outperforms others in PhD-level math and science tasks, and blind evaluators showed a clear preference for it over non-pro models over all tested categories.

So why the reduction in price. if this model is such a great asset?

The more cynical–and probably most realistic–answer is that it’s all about market share. OpenAI recently claimed it reached $10BN in annual revenue (up from $5.5 last year) and has no shortage of investors.

Money is flowing in, and it seems like OpenAI is more than happy to eat the loss if it can expand their dominant market share.

We probably haven’t yet seen all the levers OpenAI plans to use to generate profit in future.

Right now it’s in expansion mode - growing the business and the tech, acquiring other companies (like Windsurf), working with state actors, and making itself known as a brand name.

A few weeks back, OpenAI removed legal roadblocks in its Microsoft partnership that would allow it to do an IPO.

And you best believe that when it finally goes public, Altman wants to do it big.

It wouldn’t surprise me if they aimed for (and succeeded) in achieving one of the biggest (or possibly the biggest) IPO valuation in history.

Look at, for example, Amazon, who weren’t making much if any profit on their retail business when they went public – their MO was (and still is) constant growth and reinvestment of revenue to increase share price.

And this was before AWS – by far the biggest profit generator for Amazon.

The question for future users, is what does happen when OpenAI turns its eye to opening more profit channels – will service change meaningfully?

Time will tell, but for now, the o3-pro is likely to attract many new users.

There are some limitations though: temporary chats are currently not available, image generation is not supported, nor is Canvas integration.

ChatGPT pro and team users will have immediate access to the o3-pro, with edu and enterprise customers getting it next week.

Who's being hired for Meta's new Superintelligence team and what will they do?

Everything else in AI

Speaking of lawsuits... SAG-AFTRA is (tentatively) set to reach a deal with major video game companies, ending months of strikes and litigation centred on the use of AI likenesses.

Mistral AI launched Mistral Compute "Frontier AI for everyone", a Platform-as-a-Service endeavour to give developers access to AI infrastructure, including bare-metal servers, GPUs, orchestration etc. in a bid to democratize AI development.

Google added a 'fast' mode to Veo3, focused on generating 'viral' eight-second clips using text to video.

Apple announced updates to its on-device and server foundation AI models, giving developers direct access to its foundational models.

Meta's new V-JEPA 2 model boasts impressive abilities in spatial reasoning and physics, designed to help AI agents understand the world around them.

COMMUNITY

How to automate the boring stuff with n8n

Zapier is so last year.

Now, with n8n you can build powerful AI-based automations from a simple drag-and-drop interface – letting you do anything from manage your calendar, automatically send emails or slack messages, or create Jira tickets.

Or you can create a chatbot to do it all for you.

Want to summon ChatGPT to generate a polite reply when someone schedules a meeting that could’ve been an email?

How about a bot that detects when it’s 5:55 PM on a Friday and trigger an AI agent to gracefully decline all new Slack messages until Monday?

In our course n8n Automation: Building AI-Powered Workflows you can learn how to do just that.

In the space of an hour, AI engineer Maksym Mitusov will show you how to build an AI agent and create smart workflows that save you hours of drudgery. 

Sounds good, right? If you want to learn more – you can sign up right here.

Join newline Pro and get access to all our premium articles, courses and books