Google offers a proper AI intro indeed

• Below is the latest edition of w3w, my free newsletter about emerging technology, including AI, Bitcoin, Ethereum, and more . If you would like to receive it in your inbox every Sunday, please subscribe here.
• I used several AI apps to assist production of this edition of w3w. Final edit 100% by me. For fuller detail, see the newsletter's commit history on GitHub.

I generally find trial and error work pretty well when learning new technologies, especially at the beginner stage. Just use the thing for something that's low stakes (key qualifier), futz around a bit, and see what happens.

That had certainly been my approach with AI. But last week I decided to get a little more formal with things, so I earned my Google AI Essentials certification on Coursera.

I picked this one based on many positive press reviews elsewhere, as well as good word of mouth. And I'm glad to echo those kudos based on firsthand experience.

AI Essentials definitely gives a good high-level, non-technical introduction that should be useful to a lot of people. Plus, it's relatively short (perhaps a week or two of work, depending on your own pacing) and affordable at $49 for a monthly Coursera subscription.

This course won't make you a leading expert, by any stretch. But if you're an educator, journalist, small-business owner, or other professional looking for an entry point to figure out what all the AI fuss in the news is about, I can definitely recommend AI Essentials as a good place to start.

For me, as someone who had been experimenting with apps like ChatGPT and Microsoft Copilot on my own for some time, AI Essentials reinforced some things I was already doing with AI. But it also gave me a better understanding of why those things work, habits I should change, and some new ones I should adopt.

For example, I vaguely understood that AI could assume a persona in generating output from a prompt. But I had never really used that tactic much in writing prompts. Now I can defintely see how that's a useful tool to use more often.

AI Essentials also spurred me to make some tweaks my regular AI-assisted workflow for the newsletter. In this edition, ChatGPT and Google's Gemini in particular played a bigger role in selecting stories from around the web for inclusion in the news rundown below, not just drafting summaries for things I had already surfaced in my own reading throughout the week.

I can honestly say there are a few links included this time that wouldn't be here at all if not for the bots. And that's a good thing.

As ever, the final edit is 100% by me. And if you're curious to see more detail about how I use AI in drafting w3w, check out the newsletter's commit history on GitHub. 😸

Week in Review: Oct. 5-11, 2025

• A global consortium of 10 major banks, including Goldman Sachs, Citi, Deutsche Bank, and Banco Santander, is exploring development of a stablecoin pegged to a basket of major national currencies, run on a public blockchain. Analysts at JPMorgan, which is not participating in the project, estimate in a new report that stablecoins could generate $1.4 trillion in additional demand for the U.S. dollar by 2027.
• Crypto prices ran hot and cold. Bitcoin hit a fresh record above $125,000 early in the week. But then traders' appetite for speculative bets on riskier assets, including digital tokens, faded following the announcement Friday of a new round of U.S. tariffs against China. The broader crypto market's global valuation slid 9% to $3.8 trillion on fears of a prolonged trade war between two of the world's biggest economies. Gold, a traditional economic safe haven, hit a new all-time high above $4,000 per troy ounce.
• A raft of new AI product releases: OpenAI's upgraded video generator Sora 2 sparked controversy over its handling of copyrighted material, prompting the company to hastily implement new controls... At its Developer Day event on Monday, OpenAI unveiled several other products as well, including a new GPT-5 Pro language model that the company touted as a "reasoning powerhouse." The Neuron's Grant Harvey published a handy summary of all OpenAi's new stuff here... Google released a business-focused AI suite dubbed Gemini Enterprise... Meanwhile, Anthropic made a subtle but pointed critique of the glut of low-quality content generated on competiors' AI platforms by opening a "Zero Slop Zone" in New York. Visitors were urged to step away from screens and focus on human thought. The company also launched Petri, an open-source tool for automating AI safety audits.
• The tech deals just keep coming. AMD inked a chip-supply deal with OpenAI that grants the startup an equity option covering up to 10% of AMD's shares... Intercontinental Exchange, parent of the venerable New York Stock Exchange, made a strategic investment of up to $2 billion in the prediction platform Polymarket. The deal makes Polymarket’s founder Shayne Coplan, at 27 years old, the world's youngest self-made billionaire... Galaxy Digital said it has raised $460 million to convert its Texas bitcoin mining site into an AI data hub.

Market Snapshot

A quick look at some major indicators as of Friday's close on Wall Street:

Looking Ahead

• New warnings of AI-related market risk: The G20’s Financial Stability Board cautioned that heavy concentration of financial models on a few large AI providers could amplify market shocks. The body is drafting standards to monitor AI use in trading, credit, and risk management... Separtely, the Bank of England warned of a growing possibility of a "sudden correction" in global markets, specifically citing the "stretched" valuations of leading AI companies.
• But governments are also encouraging AI and crypto adoption, jockeying to keep their jurisdictions competitive in such fast-growing categories. The European Commission unveiled a €1 billion initiative to embed AI into key sectors such as energy, healthcare, and manufacturing... British regulators lifted a four-year ban on retail investors buying crypto-related ETNs, a form of publicly issued debt securities... The Reserve Bank of India said it will test digital certificates of deposit using its own official blockchain network.
• Researchers at Iran's Allameh Tabataba'i University have found that some popular large language models tend to become more risk-averse when prompted to act as women. DeepSeek and Gemini both exhibited this pattern, but OpenAI’s GPT models stayed neutral. Meta’s Llama and xAI’s Grok produced inconsistent or reversed effects depending on the prompt. (Decrypt)

Odds & Ends

• Two rare greenish comets are visible in the Northern Hemisphere this month using merely binoculars or a small telescope. (NBC News)
• Maxwell House coffee is temporarily changing its name to Maxwell Apartment to highlight the brand's value proposition in a time of rising food and housing costs. (Fast Company)

That's it for now. Thanks for reading the newsletter today! If you want to know more about w3w's history and (ahem) the author, that info is available here.

If you need to reach me directly, please email peter[at]w3w[dot]media.

Best wishes for a healthy and productive week ahead. 😊