No major large language model update this week, but don't think for a DANG second the AI world slowed down.

Nope.

It definitely didn’t.

A stable OpenAI's financial future is looking more distant than a far-off galaxy.

AI pioneers are snagging Nobel Prizes like they're going out of style.

The Department of Justice is giving Google the stink eye, and Tesla's dreaming of a world where steering wheels are as outdated as floppy disks.

(If you actually think you’re getting an AI-powered Cybercab for $30K by 2027, you might wanna ctrl+alt+delete that thought. lolz.)

But wait, there's more. (infomercial voice)

Microsoft's playing doctor with new AI tools, OpenAI's sounding the alarm on election meddling, and Apple researchers are calling BS on AI's supposed reasoning skills.

Even though no major fruits fell from the LLM trees, we still got plenty of juicy Generative AI news.

We’ll cut it to you straight, with a side of fresh fruit.

Let’s get sippin shorties.🧃

According to reports from The Information, OpenAI is facing potential losses of $44 billion and may not reaching profitability in 2029.

Yikes?

The company's spending is astronomical: approximately $7 billion annually on model training and $1.5 billion on staffing.

(Superstar engineers are really getting that NBA money.)

Reports show that current revenue stands at $2 billion from ChatGPT and $1 billion from access fees for large language models, totaling $3-4.5 billion annually.

Microsoft, a key investor with a $13 billion stake, is reportedly set to take a 20% cut of revenues.

What it means:

Don’t fall for the “LLM Company A/B/C is losing billions of dollars….. is Genertive AI dead?!? headlines.

Burning cash is part of the plan.

Right now, the LLM race is a race for users and retention. And those are races that OpenAI is winning without any real competition.

Don't panic about OpenAI's finances.

This is typical startup-to-profitability stuff. They're in a race for users, not immediate profits.

For businesses and consumers, it's prime time to leverage these tools while the getting's good. OpenAI can raise billions whenever they want – they'll figure out the mo

The world is finally giving some AI OGs their flowers.

John J. Hopfield (91) and Geoffrey Hinton (76) have been awarded the Nobel Prize in Physics for their groundbreaking AI research.

Hopfield invented the Hopfield Network in 1982, mimicking certain brain functions.

Hinton, often called the "godfather of AI," applied Hopfield's work to create networks that classify and recognize patterns in large datasets.

Without them, none of us may be reading about LLMs today or building AI-powered anything.

What it means:

AI research is probably gonna start getting the recognition it deserves, many decades later.

This is massive validation for the AI field.

It's not just tech bros playing with fancy algorithms – we're talking Nobel-level impact here. AI is changing the world in fundamental ways, and now it's got the hardware to prove it.

Is Google in some hot water that’ll impact its AI advantage?

The US Department of Justice is considering remedies to address Google's dominance in online search, which could significantly impact the tech giant's profitability and AI advancements.

Proposals include forcing Google to divest key parts of its business, such as Chrome and Android, restricting sensitive user data collection, and allowing websites to opt out of AI training.

Rut-roh.

Following the announcement, Alphabet's stock dropped 1.5%. Analysts warn these changes could weaken Google's revenue and provide opportunities for competitors like DuckDuckGo and Microsoft Bing.

What it means:

This could be big, y'all.

The DOJ's move could reshape the entire online landscape. For Google, it's a potential nightmare scenario. For the rest of us? It might mean more choice and privacy. But don't count Google out – they've got the resources to fight this tooth and nail.

Tesla is getting crushed after its AI-powered Cybercab was revealed.

Tesla unveiled its AI-powered Cyber Cab robo-taxi concept, a two-seater self-driving vehicle without traditional controls like steering wheels and pedals.

CEO Elon Musk announced plans to introduce it by 2027 with a price tag under $30,000.

However, investors were unimpressed. Tesla's shares dropped 9% immediately after the announcement and are now down more than 11%.

The company also announced plans for unsupervised full self-driving capabilities in Texas and California next year for existing Model 3 and Model Y vehicles.

What it means:

Take Tesla's promises with a mountain of salt.

(Actually, we’d advise to take anything AI-related that Elon projects with a grain of salt,

Remember the Cybertruck? Announced in 2019 at $39,000, it finally shipped in late 2023 at nearly $100,000.

Don't start saving for that $30,000 Cyber Cab just yet. It's either gonna cost 3-5 times that or we won't see it this decade.

Or both.

Can we just please make this happen already?

Whether it’s Microsoft, Google or someone else — the U.S. healthcare system is ripe to be disrupted by AI. And we’re guessing all parties involved would welcome the change.

Microsoft has announced a suite of new healthcare data and AI tools designed to alleviate the administrative burden on clinicians.

These include medical imaging models, a healthcare agent service, and an automated documentation solution for nurses.

The company has partnered with Providence Health and Services to create a whole slide model for pathology. The healthcare agent service aims to assist with tasks like answering clinical questions and identifying relevant clinical trials.

What it means:

This could be a game-changer for the healthcare industry.

Nurses currently spend up to 41% of their time on documentation – that's wild.

By automating mundane tasks, these AI tools could free up medical professionals to focus on patient care. It's about time we leverage AI to unburden our broken healthcare system.

Let’s make this happen, everyone?

Hands in on 3.

1-2-3!

Anything you read about the U.S. upcoming election?

Double check it to make sure it’s not AI-powered misinformation.

OpenAI has revealed that its AI models are increasingly being exploited to create fake content aimed at influencing elections.

In 2023, they neutralized over 20 sophisticated attempts to generate misleading articles and social media comments related to elections.

The US Department of Homeland Security has expressed growing concerns about foreign interference in the upcoming November 5th elections, with countries like Russia, Iran, and China potentially using AI to spread divisive information.

What it means:

Don’t share that on Facebook just yet, Aunt Nancy.

With the US election coming up in about 4 weeks, it's time to stay extra vigilant.

Double-check everything you read on social media or online, especially political content.

AI-generated images and videos are getting scary good, and by the time the general public realizes something's fake, it could be too late.

Shots fired.

Apple researchers have called into question the effectiveness of major large language models.

In a new research paper, they’ve developed a new evaluation tool called GSM Symbolic, which builds on the GSM8k mathematical reasoning dataset to test AI models more rigorously.

Their study tested a range of models, including Meta's Llama, Microsoft's PHI, Google's Gemma, Mistral, and OpenAI's GPT-4o and o1.

The research revealed that even top models struggle with real logic, and current accuracy scores on the GSM8k dataset are deemed unreliable. Despite high benchmark scores, OpenAI's o1 model still exhibits performance fluctuations and makes basic errors.

What it means:

This study highlights the ongoing debate about AI's true capabilities.

While companies like OpenAI claim their models can reason, Apple's researchers are saying, "Not so fast." It's worth noting that Apple's been spending big bucks trying to develop its own large language model, so there might be some competitive motivation behind this study.

As we head into 2025, expect to hear a lot more about "reasoning models" and whether AI can actually think like humans.