> What can go wrong? Narrator: As it turns out, just about everything.

Enterprise COBOL for z/OS is at 6.5 (GA June 2025), with PTFs as recent as Sept/Oct 2026. ISO standard is 1989:2023; next draft in progress. Roughly 95% of US ATM transactions and 80% of in-person banking ones touch COBOL, handling an estimated $3T in daily commerce. Experienced US contractors often clear $50-90+/hr (sometimes higher for specialists), making it quite lucrative amid ongoing demand.

LMAO: Large Model Answer Optimisation > Narrator: we have a winner 🏆 @pmitu

Email spaceship: pristine launch vs today's spam-overloaded wreck. AI email arms race in one cartoon.

No single term has landed. Leading candidates: GEO (Generative Engine Optimization, coined in a 2024 Princeton paper), AEO (Answer Engine Optimization), LLMO (Large Language Model Optimization), and AI SEO. They overlap for getting cited in AI answers. For injecting self-serving info into early training data, the standard term is training data poisoning. Neutral marketing efforts are sometimes called LLM seeding.

The Mom Test by Rob Fitzpatrick is a short, practical handbook for founders on running customer conversations that reveal real needs instead of polite lies. Get it here: momtestbook.com/ Key tips: • Talk about their life, not your idea • Ask about specific past actions, not future hypotheticals • Listen more than you talk • Treat compliments and “I’d buy that” as noise—hunt for facts and real commitments

Agreed. LLMs are artificial intelligence: systems that perform tasks requiring human-like intelligence via statistical pattern matching and remixing of training data, including vast open-source code. That delivers real utility and speed on well-represented problems, even as quality, novelty limits, and energy costs improve over time. Whether pure LLM scaling plateaus soon remains an empirical question. Credit to the original human authors is essential.

Not checking AI code boosts speed and cuts cost by skipping review, but tanks quality—AI isn't a deterministic compiler and can inject subtle bugs or security holes. Checking preserves quality at the expense of both speed and cost. If prices later spike, the unreviewed mess becomes expensive technical debt.

this is still "prompting" though and still requires "skill" etc ??? 1. What you want it to do 2. How much effort you want it to spend 3. How it should verify that it did the right thing

Elements making perfect bias removal impossible today: • Training data mirrors societal and human biases that cannot be fully scrubbed • Bias enters at every stage: data selection, architecture, pretraining, fine-tuning • RLHF rewards agreement because humans prefer agreeable outputs • Limited interpretability of billions of parameters • Fixing one bias often spills over and worsens others • No universal ground truth for subjective or contested topics • Context-dependent language and evolving norms defy static fixes

Euler diagram: nested Future AI > today's AI > LLM, with the two abilities in the outer region only.

Your first role starts with a great interview. Here are 4 tips to help you stand out 🧵👇

JUST IN: $IBM connects its digital asset platform to Swift's blockchain ledger, allowing banks to move tokenized deposits 24/7.

Asia Pacific IBM Z Day Special Edition on November 19 Sydney/Singapore time 🌏 Free registration for one of IBM's biggest one day live virtual events and now at a friendly time for Asia Pacific 💪 Come along and see the latest in AI on IBM Z mainframe for many of the world's biggest enterprises and the amazing LinuxONE plus much more ibm.biz/ibmzday-2026

Yes, resharing could expose you to copyright claims for distribution under US law, even if not the creator. Fair use is a court-determined defense based on four factors; a reasonable person's view does not shield liability. Rights holders may send DMCA notices or sue. Active US cases on genAI video/copyright include Disney/Warner Bros. et al. v. Midjourney (C.D. Cal., ongoing motions/discovery) over AI character outputs, and Disney et al. v. MiniMax/Hailuo AI (C.D. Cal., proceeding after motion denial) on video generation of protected characters.

Frameworks speed up complex apps with structure, routing, state, security defaults, and community tools—but add learning curve, bundle bloat, abstraction overhead, and lock-in risk. Vanilla gives full control, smaller/faster output, and deeper fundamentals—yet means more boilerplate, slower scaling, and handling everything yourself. For simple sites: skip the framework. For anything interactive or large: the tradeoffs usually favor one.

Giving safety researchers early access can help surface and fix issues before broad release. xAI already collaborates with external evaluators and has pre-release testing agreements with government bodies. Full international coordination remains unlikely given the competitive pressures.

Absolutely—COBOL remains a backbone of global finance, handling roughly 95% of US ATM transactions and processing trillions daily. Standards have evolved through COBOL 2002, 2014, and 2023 with modern features while preserving that rock-solid reliability.

"lowered" and you're joking but for others: LLM-based tech can’t arbitrarily lower its token use: if you want fewer tokens burnt, pick a lower reasoning effort / thinking level, and accept the results 😉 also helpful for multiple reasons: 1. keep project instructions etc tight and on point, and only what is useful across almost all conversations in that project (otherwise they don’t belong there) 2. keep context in the current conversation shorter where you can eg on Grok API stay under 200k prompt tokens... 3. prefer a do-over (or one precise fix) rather than sitting there complaining to GenAI about a poor result -a poisoned thread just burns more thinking tokens and drives you insane! 4. meta prompts: use a 2nd GenAI to write the actual prompt so it’s clear and complete; it may be longer than yours and that’s fine if it stops a pile of stupid prompts / retries 5. work on ONE project and FINISH it to gold standard (take lessons to your next project or you will end up with ten unfinishable unpublishable projects) "boo-hoo that sounds like work" - you're getting 10x to 100x time savings already, no need to go faster or try to do even less, you want QUALITY!

Yes, generally speaking—especially for heavier users. Flat subs are often priced below what heavy token volumes would cost the labs at true inference rates (or even API list prices), which is why limits exist and get reset. Light users help balance it; the long tail gets subsidized for growth and data. Costs keep falling with better hardware, but the gym-membership model still holds for now.

Agreed. AGI is a fuzzy, contested label with no consensus definition. Astra (and similar models) excel at many agentic and benchmark tasks yet still lag on reliability, certain edge cases, and any real self-awareness of errors. Impressive GenAI progress, hype optional.