How X Decides What 550 Million Users See (10-minute read)

Full disclosure: I left Twitter when it changed to X, so I’ve never used the platform.

X recently open-sourced its algorithm. This post dissects the code to find out how it works. The algorithm tries to maximize positive engagement and minimize negative engagement while prioritizing quality over quantity.

Controversial content persists despite penalties because volume overwhelms penalty. Rage bait works when the engaged audience vastly outnumbers the offended minority. Early engagement matters because of feedback loops, not because of any multiplier for fast likes.

Read the full article on Code Pointer.

Sources:
•  TLDR. (2026, February 2). TLDR Newsletter. https://a.tldrnewsletter.com/web-version?ep=1&lc=f5cb7e17-003d-11ed-9258-0241b9615763&p=f506196a-0010-11f1-8219-038103267f56&pt=campaign&t=1770033693&s=b1ae5ccaa098e95025d02e286faf4b0178ff0f03d200dff5e776f132678924d7
•  Yongkyun. (2026, January 28). [x-algorithm] How X Decides What 550 Million Users See. Code Pointer. https://codepointer.substack.com/p/x-algorithm-how-x-decides-what-550

Loyalty Is Dead In Tech (3-minute read)

It used to be that early employees who took the risk along with the founders shared the outcome. That is no longer the case, with founders leaving companies and employees in acquihires that license the technology and abandon the rest.

Employees are starting to learn that a ‘mission’ is just a recruiting pitch. There are still some founders who would rather go slower than abandon the people who believed in them — finding them has never mattered more.

Read the full blog post on Balancing Act.

Sources:
•  TLDR. (2026, February 2). TLDR Newsletter. https://a.tldrnewsletter.com/web-version?ep=1&lc=f5cb7e17-003d-11ed-9258-0241b9615763&p=f506196a-0010-11f1-8219-038103267f56&pt=campaign&t=1770033693&s=b1ae5ccaa098e95025d02e286faf4b0178ff0f03d200dff5e776f132678924d7
•  Kothari, N. (2026, January 30). Loyalty is dead in tech. Balancing Act. https://writing.nikunjk.com/p/loyalty-is-dead-in-tech

The “AI-Generated Hit Movie” Horror Story (9-minute read)

Roku’s CEO predicts that the first 100% AI-generated hit movie will be released in the next three years. AI-generated video is already flooding social media, so it is not unreasonable to believe that fully AI-generated movies will be possible in three years.

The bigger question is whether there will be an audience for that content. It is likely that AI-generated content will raise the demand for human-generated content. There may be a market for hybrid content, but a purely AI-generated movie topping the box office by 2029 is unlikely.

Read the full article on Spyglass.

Sources:
•  TLDR. (2026, January 12). TLDR Newsletter. https://a.tldrnewsletter.com/web-version?ep=1&lc=f5cb7e17-003d-11ed-9258-0241b9615763&p=75705fcc-ef98-11f0-9f88-870dc8dd9c2c&pt=campaign&t=1768219211&s=26832720481d3119de8fb7f3ab2dec03f98096d6016c3a53b6b7a939a8724c9a
•  Siegler, M. (2026, January 10). The “AI-Generated Hit Movie” Horror Story. Spyglass. https://spyglass.org/ai-generated-hit-movies

Google says it is ‘gradually rolling out’ option to change your @gmail.com address (4-minute read)

Google is gradually rolling out the option to change a Gmail address to a new one (up to three times) with an altered username. Users’ original email address will still receive emails and work for sign-in. Users will be unable to change or delete their email within 12 months of each change.

The support page detailing the update is currently only in Hindi.

Read the full article on 9T🕘5Google.

Sources:
•  TLDR. (2025, December 18). TLDR Newsletter. https://a.tldrnewsletter.com/web-version?ep=1&lc=f5cb7e17-003d-11ed-9258-0241b9615763&p=e3e9d6a2-e22a-11f0-afef-355d711e4cd9&pt=campaign&t=1766750425&s=58b821c1851cf90164444bda52a229e71373eca7b07bd111e21da74b821d4f04
•  Schoon, B. (2025, December 25). Google says it is ‘gradually rolling out’ option to change your @gmail.com address. 9to5Google. https://9to5google.com/2025/12/24/google-change-gmail-addresses/

Meta is about to start grading workers on their AI skills (3-minute read)

Meta plans to tie employee performance to their “AI-driven impact” starting next year. The company will assess employees on how they use AI to deliver results and build tools that can move the needle in a major way in regards to productivity.

Meta is rolling out an “AI Performance Assistant” to help with performance reviews and feedback to use in this year’s performance review cycle. Employees can also use Meta’s internal assistant, Metamate, as well as Google Gemini, for their performance content.

Read the article on AOL.

Sources:
•  TLDR. (2025, November 17). TLDR Newsletter. https://a.tldrnewsletter.com/web-version?ep=1&lc=f5cb7e17-003d-11ed-9258-0241b9615763&p=97c204ce-c395-11f0-87fd-89d261ee23b3&pt=campaign&t=1763380769&s=6101acfed2930ed36d57aedbe8c0ed59f740e671a80a39755dd92401b2f9001d
•  Mann, J. (2025, November 14). Meta is about to start grading workers on their AI skills. AOL. Retrieved November 17, 2025, from https://www.aol.com/articles/meta-start-grading-workers-ai-194500326.html

AI turns brain scans into full sentences and it’s eerie to say the least (8-minute read)

‘Mind captioning’ is a method developed at NTT Communication Science Laboratories in Japan that uses a blend of brain images and AI to generate text descriptions of what people are seeing (or even visualizing) based only on their neural patterns. In tests, the system was able to correctly describe scenes viewed by participants about half the time, even when presented with a hundred possibilities — and even when participants were just visualizing the scenes in their mind.

The technology is still far from being a mind-reading device, but it could have serious ethical implications when it becomes more capable.

Read the article on ZME Science.

Sources:
•  TLDR. (2025, November 7). TLDR Newsletter. https://a.tldrnewsletter.com/web-version?ep=1&lc=f5cb7e17-003d-11ed-9258-0241b9615763&p=84c5e47a-bbbc-11f0-8862-cd15a38dd90c&pt=campaign&t=1762516721&s=fe0c4ab202cbf1eee99bef73a8587f9a8fef8f2698d09a15f5bd93f204f0c732
•  Puiu, T. (2025, November 6). AI turns brain scans into full sentences And it’s eerie to say the least. ZME Science. Retrieved November 7, 2025, from https://www.zmescience.com/future/ai-turns-brain-scans-into-full-sentences-and-its-eerie-to-say-the-least/

Invaluable AI prompt advice

I continue to be impressed with the advice about AI prompts that you’re giving to create an image, which is to add a statement like “ask me clarifying questions” to the end of your prompt to help your AI program of choice create a more accurate portrayal of what you’re going for.

Recently, I wanted an image to accompany one of my exercise check-ins. I prompted Copilot AI adding the clarifying statement, and here is the back-and-forth I had with it to create the perfect image on the first try:

Draw me a photo quality image of a 50-year-old man in work casual pants walking by a company snack room holding 2 cookies on a small plate and then 4 cookies in his pocket. Ask me clarifying questions to better understand my needs. One question at a time, keep going until you’re confident you have enough to create a complete image.

Great setup—this has the makings of a wonderfully sly scene. To start shaping it visually: Would you like the man’s expression to be neutral, guilty, smug, distracted, or something else entirely?

Smug.

Perfect. Smug opens up a lot of fun possibilities—maybe a raised eyebrow, a half-smile, or even a glance toward the camera like he knows he’s getting away with something.

Next: what kind of setting do you imagine for the snack room? Is it sleek and modern, cluttered and cozy, or maybe a bit dated and corporate?

Sleek and modern.

Got it—sleek and modern snack room. Think glass-fronted fridge, minimalist counters, maybe some brushed metal and soft lighting.

Next: what kind of work casual pants is the man wearing? Are we talking chinos, slacks, tech-fabric joggers, or something else? And what color?

Tan chinos.

Excellent—tan chinos give us that classic work-casual vibe. They’ll contrast nicely with the sleek snack room.

Next detail: what kind of shirt is he wearing? Button-down, polo, sweater, or something else? And any particular color or pattern?

Button-down.

Thanks! What color or pattern is the button-down—solid blue, plaid, white with stripes, something else? And is it tucked in or untucked?

Blue plaid, tucked in.

Awesome—blue plaid button-down, tucked into tan chinos. That gives us a crisp, office-ready look with just enough personality.

Now, about the cookies: what kind are they? Chocolate chip, oatmeal raisin, snickerdoodle, something else? And are they visibly different or all the same?

Chocolate chip.

Classic choice—chocolate chip it is. That’ll make the cookie stash instantly recognizable and relatable.

Now, about the cookies in his pocket: do you want them subtly bulging through the fabric, one peeking out, or clearly visible like he’s flaunting the stash?

Mostly subtle bulging through the fabric with one sticking out a little bit.

Perfect—subtle bulge with one cookie peeking out adds just the right touch of visual mischief.

Next: do you want any other characters in the scene? For example, a coworker in the background, someone reacting, or is it just the man and the snack room?

Two other coworkers in the background reacting.

Great—two coworkers reacting in the background adds a layer of storytelling and contrast to the smug cookie thief.

How would you like them to react? Options could include:
– Shocked or disapproving
– Amused or entertained
– Whispering or gossiping
– Trying to pretend they didn’t see anything

Let me know which vibe you’re going for.

Whispering.

Perfect—two coworkers whispering in the background adds a layer of intrigue and social tension.

Next: what kind of environment surrounds the snack room? Is it part of an open office with desks and glass walls, a hallway with carpet and cubicles, or something else?

Part of an open office with desks and glass walls.

Here it is—your photo-quality image is ready now. Let me know if you’d like to tweak anything or build on this scene further.

Figure AI’s New Humanoid Robot Can Fold Your Clothes, Do the Dishes (3-minute read)

Figure 03 is a general-purpose humanoid robot that can perform a wide range of chores at home. It uses an AI-powered system called Helix to see and execute complex actions without a predetermined script. At 5 feet and 6 inches tall, the 132-pound robot can run for five hours on a single charge.

The company aims to release the robot in 2026, but acknowledges that the goal is a big push. A video from Figure AI introducing the robot is available in the article.

Read the article on PC Mag.

Sources:
•  TLDR. (2025, October 10). TLDR Newsletter. https://a.tldrnewsletter.com/web-version?ep=1&lc=f5cb7e17-003d-11ed-9258-0241b9615763&p=85bc1104-a5bb-11f0-aee3-45609736e10f&pt=campaign&t=1760093966&s=33f0f5849915b9340c183d0fb7c76f2f79ed2e35ed200df82f0b17c2477ee43f
•  Kan, M. (2025, October 9). Figure AI’s new humanoid robot can fold your clothes, do the dishes. PCMAG. https://www.pcmag.com/news/figure-ais-new-humanoid-robot-can-fold-your-clothes-do-the-dishes

Class dismissed (95-minute read)

Alpha School is a teacherless, homeworkless, K-12 private school in Austin, Texas. Its students have been testing in the top 0.1% nationally.

They learn through self-directed coursework with AI tutoring apps for two hours a day, spending the other four hours of the school day in ‘workshops’ where they learn things like how to run an Airbnb or food truck, manage a brokerage account or Broadway production, or build a business or drone.

Alpha School’s new product, Timeback, is aimed at scaling Alpha School’s concepts and results and giving students their childhoods back. Timeback will be free to everyone once it’s released.

Read the article on Colossus.

Sources:
•  TLDR. (2025, August 22). TLDR Newsletter. https://a.tldrnewsletter.com/web-version?ep=1&lc=f5cb7e17-003d-11ed-9258-0241b9615763&p=37a7efb6-7f2c-11f0-9b29-8f76d000773e&pt=campaign&t=1755860278&s=645cc1494c311c683a1eaf5cf1f3dec62eb83ea56a9d9cceb7834938948a19b0
•  Class dismissed – Colossus. (2025, August 27). Colossus. https://joincolossus.com/article/joe-liemandt-class-dismissed

GenAI FOMO has spurred businesses to light nearly $40 billion on fire (5-minute read)

US companies have invested between $35 to $40 billion in generative AI initiatives, and so far have almost nothing to show for it.

Only 5% of organizations have successfully integrated AI tools into production at scale. Confidence in AI initiatives is declining among corporate leaders.

The divide is due to the inability of AI systems to retain data, adapt, and learn over time, rather than insufficient infrastructure, learning, or talent.

Read the article on The Register.

Sources:
•  TLDR. (2025, August 20). TLDR Newsletter. https://a.tldrnewsletter.com/web-version?ep=1&lc=f5cb7e17-003d-11ed-9258-0241b9615763&p=aa1a4956-7d9f-11f0-a182-f70d4b096c94&pt=campaign&t=1755687445&s=05913a0a857d57886bd525c1b3b3375260efd1816ffe1398b07cc67a3597bef2
•  Claburn, T. (2025, August 18). GenAI FOMO has spurred businesses to light nearly $40 billion on fire. The Register. https://www.theregister.com/2025/08/18/generative_ai_zero_return_95_percent/

Another “Rhinestone Cowboy” job offer text

That’s what Bob (my husband) and I call them. Here’s the most recent one:

Who gets the “Rhinestone Cowboy” allusion?

These things are such a crock! This one was to Bob’s phone and 1) he doesn’t even have a profile on LinkedIn, and 2) he hasn’t had a resume for 40 years. (And he’s been retired for more than 10 years.)

I recently was teased with this article on Slate Plus: My Scammer: I responded to one of those spam texts from a “recruiter” — then took the job. It got weirder than I could have imagined.

I don’t have a Slate Plus subscription, but if anyone does and is willing to share access to this article, I’m dying to read it! (Although I’m not curious enough to pay for a subscription, obviously.)

Robotic neck incision replaces heart valve with no chest opening in world first (5-minute read)

Doctors at the Cleveland Clinic replaced heart valves through a small neck incision using robotic assistance, avoiding the need to open the chest.

Four patients have so far undergone the procedure, called transcervical robotic access for aortic valve replacement, and were discharged within days, with one even resuming gym workouts just a week after surgery.

The procedure is being performed exclusively at the Cleveland Clinic for now, but the team hopes to standardize it for use at other highly specialized centers. It could redefine what is possible for heart surgery.

Read the article on Interesting Engineering.

Sources:
•  TLDR. (2025, July 25). TLDR Newsletter. https://a.tldrnewsletter.com/web-version?ep=1&lc=f5cb7e17-003d-11ed-9258-0241b9615763&p=2e7b5d28-692a-11f0-b835-3db36ffea3ac&pt=campaign&t=1753441048&s=a14bf86d633d4be953fe2c165deb3428d686e017277d7ca7edd7315783c14d77
•  Walter, N. (2025, July 24). Robotic neck incision replaces heart valve with no chest opening in world first. Interesting Engineering. https://interestingengineering.com/health/robotic-heart-surgery-neck-valve-replacement

This upcoming iOS feature will make spam phone calls an issue of the past (2-minute read)

Apple announced two new features to vastly improve the phone calling experience for iPhone users this year at WWDC25: Hold Assist and Call Screening.

Hold Assist manages calls for users when placed on hold. Call Screening, designed to deter spam callers, answers calls for users and asks callers for their names and reasoning.

iOS 26 will also introduce a screening filter for texts.

Read the article on 9T🕘5Mac.

Sources:
•  TLDR. (2025, July 18). TLDR Newsletter. https://a.tldrnewsletter.com/web-version?ep=1&lc=f5cb7e17-003d-11ed-9258-0241b9615763&p=8a193296-63b1-11f0-90d6-d34c2221e311&pt=campaign&t=1752836260&s=5e637c51a3e4ebbdde5f09bb4f3fcc0cd428e519d2bc969481df13c534d12f76
•  Burkhardt, M. (2025, July 17). This upcoming iOS feature will make spam phone calls an issue of the past – 9to5Mac. 9to5Mac. https://9to5mac.com/2025/07/17/ios-26-spam-call-feature/

AI-image-generation tip

I’ve been playing with ChatGPT and AI image generation for quite a while now, but only recently did I learn what’s turned out to be a very helpful tip when creating an image:

Add these sentences to your prompt to help it create an image that’s closest to what you want and on fewer tries: “Ask me clarifying questions to better understand my needs. One question at a time, keep going until you’re confident you have enough to give a complete answer.”

Here’s an example of how that manifests. I wanted to create an image to go with my exercise check-in: “My main form of exercise is walking my plate back to the kitchen.”

My prompt:

Create an image of a person running into a kitchen and to the kitchen sink to put down their plate. Ask me clarifying questions to better understand my needs. One question at a time, keep going until you’re confident you have enough to give a complete answer.

Copilot AI follow-ups and my [responses]:

Got it, John! Let’s start shaping this visual together. First up—what kind of style are you imagining? Photorealistic, cartoony, painterly, minimalist, or something else entirely?

[Photorealistic]

Great choice—photorealistic will capture the motion and emotion with stunning clarity. Next detail: what kind of person is running into the kitchen? You can guide me on age, gender, clothing, hairstyle, expression—anything that matters for the scene.

[a 50-year-old man]

Nice! That gives us a solid start. Let’s flesh him out a bit—what’s his vibe? Is he dressed casually after work, in pajamas on a weekend morning, or maybe in athletic gear like he just came back from a jog?

[dressed casually at home after work]

Perfect—he’s just unwinding after the day. Now, let’s zoom in on the kitchen itself. Is it modern and sleek, cozy and rustic, slightly messy, or something else entirely? Think about colors, lighting, and layout too—whatever helps set the mood.

[modern and sleek]

On it! You’re about to get a crisp, photorealistic scene of a 50-year-old man, casually dressed after work, dashing into a modern, sleek kitchen to drop his plate at the sink. 🏃‍♂️🍽️ Let me know if you’d like to tweak anything once you see it.

[Make the plate dirty from unfinished food.]

Got it—your scene is about to get even more true-to-life, complete with a dirty plate showing some leftover food still clinging to it. Perfect finishing touch. Let’s bring it to life.

The image generated was exactly in line with what I was hoping for:

A complete, creepy aside:

Eerily, the image generated looks a lot like our actual kitchen sink area, complete with the white cabinets, backsplash, and faucet!

Humans Are Sounding More Like ChatGPT, New Study Suggests (3-minute read)

Humans are increasingly subtly mirroring AI language patterns by using ‘GPT words’ (common words used by ChatGPT, like ‘delve’ and ‘underscore’) in conversation.

A new study suggests humans are starting to talk more like AI, a trend that started when ChatGPT came along a couple of years ago. The study tracked the frequency of GPT words in YouTube videos and podcast episodes from before and after ChatGPT’s launch.

Its results are a reminder that AI’s influence extends beyond just tools and into the very fabric of how we communicate.

Read the article on Android Headlines.

Sources:
•  TLDR. (2025, July 17). TLDR Newsletter. https://a.tldrnewsletter.com/web-version?ep=1&lc=f5cb7e17-003d-11ed-9258-0241b9615763&p=32e65c9e-62ec-11f0-841b-29051218926f&pt=campaign&t=1752749816&s=0ef92b1c6c2f50add1f27b3ba05fec59dd45f457d8678305e548adad32622387
•  Leon, J. (2025, July 16). Humans are sounding more like ChatGPT, new study suggests. Android Headlines. https://www.androidheadlines.com/2025/07/humans-talking-like-ai-chatgpt-study.html

Windows is getting rid of the BSOD after 40 years (2-minute read)

Microsoft plans to overhaul its blue screen of death (BSOD) error message in Windows 11 in favor of a simplified black screen known as the Black Screen of Death.

Read the article on The Verge.

Sources:
•  TLDR. (2025, June 27). TLDR Newsletter. https://a.tldrnewsletter.com/web-version?ep=1&lc=f5cb7e17-003d-11ed-9258-0241b9615763&p=29c1d0c0-532a-11f0-8a05-dfd4e7565873&pt=campaign&t=1751021762&s=dc783f725385f35826ce3417523b7fb4d1e69f6f44d7ed849655265c20bdc9ab
•  Warren, T. (2025, June 26). Windows is getting rid of the Blue Screen of Death after 40 years. The Verge. https://www.theverge.com/news/692648/microsoft-bsod-black-screen-of-death-color-change-official