AI's heating up faster than a GPU farm in Death Valley today. From JPMorgan's AI performance reviews to OpenAI's maverick dealmaking team, the machines are getting cozy in our workflows.
JPMorgan Lets AI Write Your Performance Review (What Could Go Wrong?)
JPMorgan just dropped the corporate equivalent of 'here, let the robot judge your value' by allowing employees to use their internal AI system to draft performance reviews. Yep, you read that right – workers can now skip the hours of corporate-speak gymnastics and let algorithms summarize their professional worth. Unlike sneaking off to ChatGPT (which some employees were definitely doing anyway), this happens entirely within JPMorgan's secure system using their proprietary language model. The move represents one of Wall Street's boldest embraces of AI for everyday work tasks. Here's the real question: will AI reviews be more or less brutal than your actual manager? 'Jeff shows excellent potential but spent 43.7% of Q3 playing Wordle during meetings' might be too honest even for annual reviews. For JPMorgan, it's a massive efficiency play – but also a fascinating real-world test of whether AI can handle nuanced workplace politics.
🔥 This hits different, right?
Share this with someone who needs to see what's really happening in AI
Australia slapped Microsoft with a lawsuit yesterday claiming the tech giant misled 27 million customers about how it collects and uses their data, proving even AI-powered megacorps can't escape regulatory heat.
•
Qualcomm jumped into the AI chip race this morning with its new AI200 and AI250 processors, challenging Nvidia and AMD's dominance in the market that's hotter than fresh silicon.
•
OpenAI's Sora video generator is being called a 'bizarre mind-bending AI slop machine' in a Los Angeles Times piece published yesterday, igniting fresh debates about hyperrealistic AI videos as app downloads surge.
•
IBM launched a blockchain-based digital assets platform today, letting financial institutions build crypto services just as market activity is heating up again.
🔌 API WATCH
Qualcomm's AI Chip Entry Could Reshape Edge AI Economics
Qualcomm's AI200/AI250 announcement today signals a major shift in edge AI economics. While Nvidia's H100/H200 dominate data centers, Qualcomm's new chips target efficient on-device inference with specialized AI cores that could dramatically cut inference costs. Early benchmarks suggest 2-3x better performance-per-watt than comparable AMD offerings. For engineers building edge applications, this means potentially running models with 3-5B parameters locally without crushing battery life. Expect initial pricing to undercut competitors to gain market share. If you're building mobile or IoT applications that need local AI without cloud dependencies, these chips should be on your radar for 2026 deployments. The technical specs suggest particular strength for multimodal applications like Ladder's new food recognition system.
💰 FOLLOW THE MONEY
AI Infrastructure Bets Heat Up as IBM Enters Digital Assets Space
IBM's digital assets platform launch today reveals where smart money sees AI and blockchain converging. Rather than chasing consumer crypto hype, IBM's targeting institutional infrastructure – the picks and shovels of the coming tokenized economy. This follows Accel and Prosus announcing yesterday they're tag-teaming investments in early-stage Indian startups, with AI-powered fintech getting special attention. Meanwhile, tZero's 2026 IPO plans (announced this morning) suggest digital securities markets are maturing faster than expected. The pattern? Major players are betting big on the infrastructure layer rather than specific AI applications. Look for more enterprise blockchain + AI convergence plays as companies position for regulatory clarity expected by mid-2026.
🛠️ TOOL SPOTLIGHT
PACER (Panic Attack Counseling Agent)
Fresh off arXiv yesterday (2510.21143), PACER is the first AI specifically trained to talk people down from panic attacks. Unlike general therapy bots, it combines empathy with directive support based on a new dataset of first-person panic narratives. Human evaluators consistently preferred it over both generic counselors and even GPT-4 powered alternatives. Mental health tech that actually works? We're here for it.
💼 SUCCESS STORY
Ladder Fitness App
Ladder just flexed on the entire fitness app industry by launching an AI-powered nutrition tracking feature that analyzes food with just a photo snap. Announced this morning, the new tech automatically identifies what you're eating and calculates macros – eliminating the tedious food logging that makes most people quit nutrition tracking. It's one of those 'why didn't this exist already?' innovations that turns something people hate (counting calories) into something frictionless. Expect user retention numbers to pump up like they've been hitting the protein.
🚀 WANT THE EDGE THAT MOVES THE NEEDLE?
Join the AI professionals who are 10x ahead of everyone else
⚡Exclusive insider scoops
🎯Deep-dive analysis
💰Revenue opportunities
🔍AI trend predictions
🛠️Premium tool access
🧠AI-powered insights
⚡ Limited spots available for serious AI professionals
📊 BY THE NUMBERS
93%
Accuracy achieved by new LLM-powered medication extraction system across heterogeneous EHR systems, according to research published on arXiv today (2510.21027) – potentially revolutionizing how healthcare systems standardize prescription data.
51.6%
Reduction in reasoning length achieved by MedAlign framework while improving medical visual question answering performance by 11.85% compared to baseline models, according to a paper published yesterday (2510.21093).
$1.5 trillion
Total value of deals that OpenAI has navigated without traditional external advisers, using instead a small internal team assembled by Sam Altman, according to Financial Times reporting published this morning.
⚠️ IMPORTANT DISCLAIMER:
The information provided in this newsletter, including the "Follow the Money" and "API Watch" sections, is for educational and informational purposes only. It does not constitute financial advice, investment advice, trading advice, or any other sort of advice. You should not treat any of the newsletter's content as such. Promptobello does not recommend that any cryptocurrency, security, or investment should be bought, sold, or held by you. Do conduct your own due diligence and consult your financial advisor before making any investment decisions. Past performance is not indicative of future results.
That's a wrap!
Forward this to a friend who needs to stay on top of AI news.
They'll thank you (and maybe buy you coffee).
