TrendPulse

Article URL: https://www.bloomberg.com/news/articles/2021-01-12/what-do-wall-street-leaders-think-is-the-next-big-risk Comments URL: https://news.ycombinator.com/item?id=48419773 Points: 1 # Comments: 1

Article URL: https://keanw.com/2026/03/a-diary-of-an-agentic-retro-gamer-part-1.html Comments URL: https://news.ycombinator.com/item?id=48419824 Points: 1 # Comments: 0

Especially annoying since I’m a paying customer

I need help . pls help ! Actually im beginner to Ai /ML , i started Langchain , RAG idk if i'm going right or wrong ,its like im just going with flow and felt like im just learning a wrapper of real model . can anyone suggest me any course, advice,roadmap for Ai/ML , i saw andrej karpathy’s channel looks lil complex to me and saw many yt channels on ML , LLM AND THOSE WERE BASIC , like structure of ANN, RNN etc i want pure depth knowledge [lin...

Does anyone else really hate it when you ask ChatGPT for help for something, you do something wrong and you call ChatGPT out for giving wrong instructions, and they go "You're right, that's completely on me"? I hate it so much, especially with the image generation. I love anthropomorphic cartoon animals so whenever I want it to wear an accessory like gloves, it gets flagged for "third-party content". Or if I want a closeup of his mouth while he's smiling, it gets flagged fo...

The writing seems to be on the wall: long term memory will be phased out for a more fluid ever evolving ... whatever. For a lot of people, context carrying over is actually unnecessary and I agree that in doing research related tasks this can be a major annoyance at times. But I personally use gpt to mess around with characters in a fictitious cartoony world. Since the moment I even noticed long term memory was a thing it was such a helpful tool to help shape this world, so scripts and suc...

An issue caused some user accounts to be incorrectly suspended. We’re restoring access and working through related subscription and credit issues. https://status.openai.com/incidents/ejj40mae

Hi internet friends, I recorded a workshop about building your own LLM without any math / ML prerequisites. By the end of the workshop people have their own working OpenAI GPT2-style transformer, which hopefully makes it relevant to this sub. The workshop covers everything from machine learning fundamentals, deep neural networks, transformer architecture, and pre/post-training. The only prerequisite is being comfortable with learning through code & excel examples. Sampling Large Lan...

Found an AI account on instagram that is using a REAL persons face/identity. They even have more followers than the real account and have been posting for over a year. The rabbit hole of this account will lead you to see they are doing this to multiple REAL people, and likely using them to make money. I can understand AI accounts, but why use an actual persons face/identity?

Disclosure: I work on the benchmark below, so flagging that up front. We've been testing whether LLMs can critique recent science-paper summaries — catch planted flaws, overclaims, and missing evidence — and, separately, how calibrated they are about their own judgments (confidence scored with Brier, a strictly proper rule). The pattern that keeps showing up: the models best at spotting problems are also among the most confidently wrong when they miss. Critique skill and calibration look l...

Is this significant news?

Who governs AI agents once they're running in production? I went looking for the answer. It's more complicated than the press releases suggest. This week Cognizant and ServiceNow announced a partnership specifically to close what they're calling the "enforcement gap" in enterprise AI governance. The Everest Group analyst quote from the press release cuts to it: "The hard part of AI governance was never writing the policy. It's enforcing it as systems learn and act." H...

It's pretty annoying that Codex uses agents.md and Claude Code uses Claude.md. There should be some industry standards to this stuff?

I’ve been noticing a clear split lately. The big mainstream models are getting more and more restricted with heavy safety rules, while at the same time more people are switching to local or less restricted models because they actually let you explore ideas freely. It feels like we’re heading toward two different types of AI: one that’s heavily controlled and "safe", and another that’s more open and unrestricted. Both seem to be growing at the same time. Do you think this divide wil...

been building AI agents for a while and noticing a pattern: the LLM reasoning part works. the part that breaks is everything around accounts, logins, and verification. agent gets to "sign up for this service" and then: - email verification loop breaks - OTP times out while the agent is mid-step - captcha or bot detection fires - session expires between steps the model figured out what to do. the infrastructure around it didn't cooperate. curious if this matches what other...

Back in the 1980s a debate raged about whether it was okay to let children use calculators in elementary school. Critics warned that giving kids calculators would lead to the "destruction of student math skills." A similar debate is happening today across a range of areas, including coding, writing and even music. Will using AI lead a brain drain across these and many other areas? One of my favorite authors is Isaac Asimov. He's better known for his Foundation and Robot series of ...

After spending the last few weeks reading through the reasoning literature, I noticed a trend that seems worth discussing. For the past 2–3 years, a large fraction of progress in LLM reasoning came from making models generate more intermediate thoughts. Chain-of-Thought prompting (Wei et al., 2022) pushed PaLM 540B from roughly 18% to 58% on GSM8K. Self-Consistency added another 17.9 percentage points by exploring multiple reasoning paths before committing to an answer. Tree-of-Thoughts la...

I am 19, and the Founder and CEO of AutoFlow. I want to be entirely transparent before discussing our current team or your potential role: you should know exactly the engineering challenge we are tackling. We are building the trust infrastructure for the AI era. The Problem: Today’s AI systems are powerful but fundamentally unreliable for enterprise-scale execution because they rely too heavily on probabilistic pattern prediction. Responses vary, context rot sets in over long tasks, halluc...

I’ve been reading through OpenAI’s Codex Sites docs, and my takeaway is that this is not really “another AI website builder.” It feels more like Codex getting a deployable surface. The important part is not that it can generate a page. Lots of tools can do that now. The interesting part is the loop: Prompt → code → preview → save version → deploy → shareable URL → workspace permissions That changes the role of Codex a bit. Instead of only being a coding assistant that edits files or ...

Dreaming: Better Memory for a More Helpful ChatGPT

Is this what living in a golden age feels like?

ok so i was literally 3 prompts away from a finished project. i could SEE it. and then chatgpt just. starts forgetting things. first the file struture. then what we were even building. then it confidntly suggests something we already tried and shot down an hour ago like it was a genius new idea. then it goes "i think we should start fresh in a new conversation" BROOOOOO. i have been explaining this codebase to you for 4 hours. we are NOT starting over...separated everything out after t...