Jesse Manning

Alison Luchs, who has worked at the National Gallery of Art for 47 years, agreed to learn Gen Z slang and make videos because she wanted to raise interest in the museum’s art. She never expected to slay. wapo.st/4qIqlTl

@united I’m normally pretty understanding of delays, but your continued avoidable issues today are trying my patience 😖

@LeadToday Then let that be the criticism. We put Joe the Plumber on a pedestal in 2008 because he was “so representative of the common man.” Guy didn’t actually know much of anything, but I can’t imagine ever trashing him because of his work. People who work deserve respect for working imo

If she had taken a little time to learn some things about Congress, how businesses work, a little about law enforcement. Anything really. But a few days into her first term she starts talking about how to regulate businesses and “spreading around the wealth” to people who choose not to work. She’s a knucklehead of the worst kind, willfully ignorant by choice.

Well, they elected a bartender to Congress so…

@Robertson1221 @dcr63020 @FluteMagician No, it’s the age, the inability to empathize with the average person, and the inability to understand how the world has changed. Pipes and queens don’t have any power in the modern world. Let them grow old, but keep the elderly out of the U.S. government.

@dcr63020 @FluteMagician I don't agree about the age. Pope Leo is 69. Queen Elizabeth II died at 96 and was in great form well into her 80's and early 90's. Eisenhower was 70 when he finished his term of office. It's the individual, not the age.

1) MAGA influencers are trying to cover up Trump’s decline. This photo is not from today or yesterday. It’s from January. But there’s a reason this photo was chosen. See next post⬇️

@grahammorgan101 @realGulDukat Yes, in 2375 that was true. In 3190? Who knows. Odo’s influence on the link could have dramatically changed things and led to releasing the surviving Jem’Hadar from servitude and genetic engineering to make them a self-sustaining species. It’s an 815 year difference.

@realGulDukat Aren't they bred in tubes and have a genetically dependent drug fixation? I thought they were all male because it made a nastier fighting force.

ATTENTION BAJORAN WORKERS. If you are upset about a female Jem'Hadar in the 32nd Century, you understood the Jem'Hadar storyline like a Pakled. This is the logical progression from where we left them & the Dominion. THANK YOU FOR YOUR ATTENTION TO THIS MATTER. #StarTrek #Sisko197

@owl_elc @weswinder @mazeincoding I still use em dashes frequently, and “DuH Ai WrOtE tHiS fOr YoU” is the new lazy insult for idiots.

@weswinder @mazeincoding People shouldn't let AI push them around. Em dashes serve a purpose. I'll continue to use it and if people think I'm using AI there's really nothing I can do about that.

just got rejected from yc for using all lowercase in our application

I’ve done a lot with AI over the last couple of years, but this might be my favorite @JimNorton @AnthonyCumia

Cursor added Claude 4 Sonnet, and I don't think I've ever been more frustrated by an AI model. Lots of weird loops and independent actions that broke functionality. On the plus side -- it would use cuss words in its responses, mirroring the exact same energy I was giving it.

Ronald Reagan on Tariffs youtube.com/shorts/Mj6N-WB… via @YouTube

@ben_gym18 @EdKrassen Innocent casualties of a system like this seem like statistical footnotes until you become one of those innocent casualties.

Criminals thrive on the indulgence of society’s understanding. Why do you think Japan has a 99.8% conviction rate? They get 23 days to hold the accused so they can bully them into confession which is an instant conviction. They understand if they strike fear into criminals by convicting nearly everyone, even some innocent people, they create a safer society. This is why Japan is one of the safest countries in the world.

We can—and must—do both: demand swift, sure justice for Rachel Morin, and also defend due process for every person in America, citizen or not.

Your inbox is broken. Eric Vaughn is fixing it—with AI. He turned a legacy enterprise software company into an AI startup and built a tool called Eloquence that does what Superhuman and Gmail won’t: ✅ Auto-replies to inbound emails (in 5 mins or less) ✅ Feels human—complete with name, persona, even social profiles ✅ Speaks 160 languages ✅ Loops in real humans when needed ✅ Learns from your answers ✅ Doesn’t hallucinate I sat down with him to talk AI agents, email overload, corporate fear, and what it really takes to build an AI-first business. Eric @TheGenAICEO is founder of ignitetech.ai

@AnthonyCumia Oof, you’re not wrong. And no way to completely turn it off, either. I forgot to turn it off one night and the engine shut down at a busy intersection. That was when the battery decided to die. Ridiculously unnecessary experience.

The single most infuriating “feature” added to automobiles since the crank start. It’s not only annoying as hell it’s dangerous. Do away with this garbage.

@slotkinjr @danshipper @ChatGPTapp @every Same for me. I don't see any difference in image quality so far -- asking for photorealism is still getting me the DALL-E-style cartoon people.

@danshipper @ChatGPTapp @every dan-what am I doing wrong? I’m trying to make things like you made and I’m getting the same old output I used to get with bad text and bad consistency, etc.. My app appears to be updated and I’m in 4o

OpenAI just launched native image generation in @ChatGPTapp and it is INSANE We've been testing it internally @every for a few weeks—and it's by far the best image model I've tried: 1. It's incredibly good at text. It can even write out whole paragraphs without a problem. 2. It follows instructions well. You can ask it to modify small parts of an image and it will reliably reproduce the image with the modifications you requested. 3. It's good at capturing style. If you give it a reference, it'll reliably help you get the vibe.

The most fascinating stat from this survey: "41% of Millennial and Gen Z employees confess to sabotaging their company's AI strategy by refusing to use AI tools or outputs." My guess is that there's A) poor communication from leadership on how these tools should be used, leading to these generations seeing them pushed as replacements rather than supplements, and B) a genuine fear that AI is going to create severe disruption for their futures. Regarding point B, they're not wrong -- it will create disruption, and instead of fighting it, the people who are best prepared to thrive in an AI future are those who embrace it and understand it now. But the fear is real, which is why addressing point A is so important. I'm technically a Millennial. For us and Gen Z, the job market isn't great. The housing market is abysmal. Things that used to be big investments, like TVs or computers, are cheap and the necessities make life a genuine struggle. I think responsible executives are the ones who embrace an AI-first mindset that frames how it should be used to supplement an individual's work, as a partner that helps to produce more or work faster, and invest in educating your workforce so that when those inevitable disruptions do hit in full force, they're prepared. axios.com/2025/03/18/ent…

We do need to be talking a lot more about this. Today -- for the most part -- AI can be implemented as a supplement, a partner for people to increase their efficiency to either produce more, produce faster, or both. As generative AI continues to develop and agents become commonplace, entire careers will be irrevocably and massive amounts of people are at risk of being completely replaced. If someone with no coding skill can create a simple app or website today with Cursor, where will that technology be in five years? Ten years? It's already a tough job market for knowledge workers. What happens when AI truly can replace people -- one agent doing the work of 10 people, 24/7 and for no pay or benefits? Government ought to be talking about: 1. How do we help people whose entire careers have been upended by AI? How do we support them if jobs for their skill sets simply no longer exists? 2. How can we quickly upskill and assist those who are just starting their careers in fields that will be completely disrupted by AI? 3. If AI supplements the work normally done by entry-level / younger workers in a particular field and is guided by far fewer experienced workers, what do we do when those workers start to retire? How can we keep a skilled workforce on deck for when they are needed? I don't see any governments taking the inevitable and significant disruption seriously right now.

@PalePen18 @itseriknagel Jack Sub-Paar

@itseriknagel Everyone nominate Scorch

The Radio HOF is allowing public nominations until March 31. Fill out a nomination for Opie & Anthony and even one for Jim Norton. radiohalloffame.com/nominate

Am I the only one who was not particularly impressed with this launch video? I know the trend is to lose our collective minds when a new Chinese AI product is released, but I was doing the same resume reviews with a custom GPT over a year ago. The real-estate search results were well-formatted but didn't appear to be anything that the o1 model couldn't do. And while pushing a website with interactive charts to the web is nice, it didn't seem particularly revolutionary ... maybe it's impressive because of the Cursor-like functionality built right into the LLM interface? I know this industry is changing at warp speed, with new and significant innovations rolled out every week (at least ... sometimes it's every day). But why are AI influencers losing their minds at launches like this? There doesn't seem to be anything revolutionary here. Yet. youtube.com/watch?v=K27diM…

@rowancheung Please share the prompt! I’d love to see how to get such a complex result.

Anthropic's just dropped Claude 3.7 Sonnet, the best coding AI model in the world. I was an early tester, and it blew my mind. It created this Minecraft clone in one prompt, and made it instantly playable in artifacts:

@andrewhart @1x_tech I made this same mistake till Norm MacDonald corrected a talk show host once upon a time: adjective: non-plussed 1. (of a person) surprised and confused so much that they are unsure how to react. "He would be completely nonplussed and embarrassed at the idea"

@1x_tech Video is great. It creates the illusion that this is a general-purpose humanoid, able to casually carry out many misc tasks. Plus the 'soft' form factor + actors non-plussed by its presence. But unless some dramatic AGI breakthrough, I think it's an illusion.

Introducing NEO Gamma. Another step closer to home.