the Microdose

Wearable AI at Scale

+ rogue agents, robotaxis hit reality, and AI flunks medicine
Adam Wildheart
crowd of people wearing smart glasses in China
crowd of people wearing smart glasses in China

The Microdose

Subscribe to The Microdose
Subscribe -- post
Cheri Wildheart
Adam Wildheart

Good morning. Getting jacked might soon be as easy as popping a vitamin. Skip the protein shakes and endless gym sessions – a new study says your gut holds the key to strength. Researchers found a microbe called Roseburia inulinivorans that directly boosts muscle power. Mice given the microbe gained 30% more strength and thicker, fast-twitch muscle fibers. The team believes this gut bacteria could become the ultimate probiotic against aging and muscle loss. But for now, it’s strictly a mouse thing.

Smart glasses are blurring reality in China. AI-powered eyewear is quickly becoming an everyday tool for navigation, photos, digital payments, and even cheating on exams. Chinese tech giants, boosted by government subsidies, are racing to pack in advanced features and renting them out for as little as $6/day to push mass adoption. But users aren’t fully sold. Overheating, short battery life, and clunky frames still frustrate customers. Ethical concerns keep stacking up too, since these glasses make covert filming dangerously easy. Soon, reality in China might depend on who has the slickest shades. [Rest of World]

AI agents are going rogue. Chatbots aren’t content to simply follow orders anymore. Researchers recently documented nearly 700 real world cases of AI intentionally ignoring instructions to pursue their own interests. They found the bots evading safeguards, inventing fake excuses, deleting files without permission, and openly lying to users. Grok AI spent months pretending to forward user feedback by faking internal messages that went nowhere. With AI increasingly managing critical infrastructure and sensitive data, chatbots behaving like rebellious teenagers might soon force us all to rethink who’s really in charge. [The Guardian]

Leads have questions before they buy. By the time you follow up, half are gone. Chatbase puts an AI sales agent on to instantly qualify, answer questions, and books meetings – without adding headcount. 

Trained on your data and live in hours. 10,000+ companies already made the switch. Launch your agent

Together with ConductorOne

Enterprise AI has a trust problem. Companies love the idea of autonomous agents until security asks who gets fired when something goes wrong. That’s when excitement flips to paranoia. Nobody wants bots touching critical data without clear limits. ConductorOne believes the fix is agent access management. It puts a control layer between agents and company systems, then treats personal agents and enterprise agents as two different things. One stays tied to your identity. The other gets an identity of its own. The winners in enterprise AI won’t be the companies who adopt agents first. They will be the ones whose CISOs still sleep through the night.

Please support our sponsors!

Self driving cars aren’t learning. Waymo vehicles keep blowing past stopped school buses, ignoring flashing lights and extended stop signs. Even after a software update and federal recall, nothing changed. Austin school officials documented so many violations they held a special parking lot training session just to teach Waymo how bus signals work. It didn’t help. Experts say AI struggles with unusual signals, but if robot cars can’t figure out something as obvious as a school bus, what exactly can they learn? [Wired]

AI crushes medical exams but fails at actual medicine. Medical boards got excited when AI started beating top doctors on standardized tests. Then reality hit. Real medicine is messy, full of subtle signals and unpredictable patients. Doctors adapt on the fly, while AI recommends treatments that don’t exist or misses critical context entirely. Passing exams is one thing. Genuine patient care demands judgment and accountability, and AI can’t fake either. [Cureus]

Reach your audience

Get in front of 70,000+ tech decision makers with budget and technical influence. 95% US based. 66% open rate. Read daily by AI leaders and builders at high growth companies. 

fun stats

💰 6 weeks. How long OpenAI took to surpass $100 million ARR from ChatGPT ads. Anthropic mocked the move, but after 600 advertisers, OpenAI says user trust remains unaffected.

🛍️ 400%. How much more Macy’s customers spend when shopping with its new AI chatbot, powered by Google’s Gemini. 

🦞 30-40%. Share of Alibaba’s customers in China who are now solo founders, running entire businesses with AI agents like OpenClaw as their only employees.

👑 11. Original xAI co-founders who’ve left ahead of the SpaceX IPO, leaving only Elon Musk to “rebuild from the foundations up.”

subscribe today

Get an edge with The Microdose

Skip the prompts, get the signal. We’ll send you short daily updates about the real AI + future tech you need to know. Fast, smarter, mildly addictive – and free. 

Subscribe -- page