We're Doomed

Everything is Fine

Author

AI For All
February 28, 2025

In partnership with

Hello readers,

Welcome to the AI For All newsletter! OpenAI released a research preview of GPT-4.5, Anthropic announced Claude 3.7 Sonnet and Claude Code, Amazon introduced Alexa+, and NVIDIA prevailed against the DeepSeek fallout. Am I ok? Let’s find out!

Everything is Fine

Me trying to repress my negativity

Anthropic announced Claude 3.7 Sonnet and Claude Code. Sonnet is a hybrid reasoning model, which means that it combines an ordinary LLM and a reasoning model into one unified experience. It’s essentially what OpenAI wants to do with GPT-5. Speaking of which, GPT-4.5 was released. GPT-4.5? Why so tentatively named?

According to The Information: “A person who’s tested the model told us that its performance on certain tasks have been mixed; for instance, Anthropic’s recently released Claude 3.7 Sonnet beats it on certain benchmarks […] Perhaps we shouldn’t be surprised. In November, we broke the news that OpenAI had been seeing slowing improvements from training new models.” [screenshot]

I’m sure this is just a minor, easily surmountable setback. Quick, look at Claude Code! This command line tool for “agentic coding” can search and read code, edit files, write and run tests, and commit and push code to GitHub! That’s Claude Code! Did I mention it works with code? Well, it does — in limited research preview! 🥳

In other news, Amazon is rolling out Alexa+, the AI-powered version of Alexa, to select Echo devices. The LLM driving Alexa+ relies on a new architecture to connect to thousands of services and devices, with Amazon emphasizing reliability and accuracy. How reliable and accurate Alexa+ truly is remains to be seen. Alexa has only ever lost Amazon money. Needless to say, generative AI won’t change that.

There are those who will tell you that revenue from AI is meager and profits are at zero. Even so, Big Tech is spending $320 billion on data centers and AI infrastructure in 2025. Now, you may question this move, but I implore you to consider the thousands of jobs that these data centers will create. Pay no mind to the fact that these data centers won’t create thousands of jobs. That’s not relevant. Would you have it be said that American tech companies can’t do a proper job? Don’t you think it’s important to show China that they can’t break us in body or in spirit? Besides, NVIDIA proved that demand for its AI chips is still hot (for now) post-DeepSeek. You see, everything is fine.

Random unrelated image of money being lit on fire

Admittedly, this week was not without its red flags. Salesforce forecasted revenue below Wall Street expectations amid slower adoption of its Agentforce platform. “Given how poor initial generative AI experiments were for many companies, they're not just writing blank checks until Salesforce shows them Agentforce actually works,” said industry analyst Rebecca Wettemann. Additionally, Microsoft canceled some leases for AI data centers, and Airbnb CEO Brian Chesky said of AI coding assistants, “I’ve talked to a lot of other tech CEOs. Most of them haven’t seen a material change in engineering productivity.” Nothing to see here. Ok, we’re ending the newsletter right now.

🔥 Rapid Fire

Receive Honest News Today

Join over 4 million Americans who start their day with 1440 – your daily digest for unbiased, fact-centric news. From politics to sports, we cover it all by analyzing over 100 sources. Our concise, 5-minute read lands in your inbox each morning at no cost. Experience news without the noise; let 1440 help you make up your own mind. Sign up now and invite your friends and family to be part of the informed.

🔍️ Industry Insights

“As AI technologies are deployed in real-world networks, it's important to ensure that models work robustly in diverse environments. Training datasets should be sufficiently diverse for AI models to learn effectively; however, it's unrealistic for them to cover all possible scenarios. Thus, it's critical for AI models to generalize their training to handle new situations.”

Source: Qualcomm

💻️ Tools & Platforms