Chatgpt

entities · 64 notes linked

Related: Openai · Large Language Models · Prompt Engineering · AI Safety · GPT-4 · Anthropic · Generative AI · Google

Notes

• 'If journalism is going up in smoke, I might as well get high off the fumes': confessions of a chatbot helper — Human annotators writing gold-standard training data for LLMs
• 255 — Paul Christiano's defense of RLHF research positive net impact
• 27 AI models were ranked by the public and ChatGPT came 8th — these are the models that beat it — Prolific's Humaine leaderboard ranks Gemini first in public user experience study
• A Coder Considers the Waning Days of the Craft — Programmer's first-person reflection on GPT-4 displacing coding craft
• A Google researcher – who said she was fired after pointing out biases in AI – says companies won't 'self-regulate' because of the AI 'gold rush' — Timnit Gebru warns AI gold rush prevents corporate self-regulation on bias
• A former Google researcher behind a seminal AI paper describes how the company lost a top chatbot visionary — Google's reputational caution caused it to miss chatbot opportunity
• AI Is a Waste of Time — AI tools as entertainment and time-wasting before productivity gains materialize
• AI Prompt Engineering Is Dead — Automated prompt optimization outperforms human prompt engineering
• Amazon warns employees not to share confidential information with ChatGPT after seeing cases where its answer 'closely matches existing material' from inside the company — Amazon restricts employee ChatGPT use over corporate data leakage risk
• Artificial General Intelligence Is Already Here | NOEMA — Argument that current frontier LLMs have already achieved meaningful artificial general intelligence
• At TED AI 2023, experts debate whether we've created "the new electricity — TED AI 2023 first AI-only TED conference debates AGI benefits and risks
• Black Hat 2023 Review: LLMs Everywhere | DNSFilter — Black Hat 2023 LLM applications in security; best use leverages embeddings not chat interfaces
• Can AI really be protected from text-based attacks? | TechCrunch — LLM prompt injection attacks are low-barrier and currently unpreventable
• ChatGPT Gets Its "Wolfram Superpowers"! — ChatGPT plugin connecting to Wolfram Alpha and Language
• ChatGPT Makes OK Clinical Decisions—Usually — Study finds ChatGPT 72% accurate across clinical decision tasks using fictional vignettes
• ChatGPT is 'not particularly innovative,' and 'nothing revolutionary', says Meta's chief AI scientist — Yann LeCun argues ChatGPT is solid engineering not scientific breakthrough
• ChatGPT is not all you need. A State of the Art Review of large Generative AI models — Taxonomy of 2022-2023 generative AI models across all major modalities
• Chegg's stock plunges on fears of competition from ChatGPT — ChatGPT disrupts edtech as Chegg stock crashes 48% in one day
• Design and Code with ChatGPT and Midjourney — Course building apps with GPT-4 and Midjourney
• Early Evidence of Vibe-Proving with Consumer LLMs: A Case Study on Spectral Region Characterization with ChatGPT-5.2 (Thinking) — LLM-assisted iterative mathematical proof via generate-referee-repair pipeline
• Employment for computer programmers in the U.S. has plummeted to its lowest level since 1980—years before the internet existed | Fortune — AI correlates with historic drop in US programming jobs
• Even experts are surprised by AI's latest 'vibe-mathing' advance — Amateur uses GPT-5.4 Pro to solve 60-year-old Erdős primitive sets problem
• Fine-tuning ChatGPT: Surpassing GPT-4 Summarization Performance — 63% Cost Reduction and 11x Speed Enhancement using Synthetic Data and LangSmith — Fine-tuned ChatGPT beats GPT-4 summarization at 63% lower cost and 11x faster using chain-of-density synthetic data
• For Some Autistic People, ChatGPT Is a Lifeline — Autistic people using ChatGPT for social scripting and communication support
• Forget ChatGPT Canvas — I just tried Gemini Canvas and I'm floored by the difference — Gemini Canvas outperforms ChatGPT Canvas for AI-assisted writing feedback
• GitHub - darrenburns/elia: A snappy, keyboard-centric terminal user interface for interacting with large language models. — Keyboard-focused terminal app for chatting with multiple LLM providers
• Github Copilot and ChatGPT alternatives — Survey of AI coding assistants and privacy-safe ChatGPT alternatives in 2023
• Godfather of Artificial Intelligence" Geoffrey Hinton on the promise, risks of advanced AI — Geoffrey Hinton 60 Minutes interview warning of AI existential and societal risks
• Harvard '21 grad says Gen Z just uses A.I. to do their homework — Gen Z uses ChatGPT for homework completion rather than genuine learning
• How ChatGPT is changing the way cybersecurity practitioners look at the potential of AI — ChatGPT dual-use cybersecurity capabilities surprise skeptical security researchers
• How Claude Helps Me Manage My Calendar (but ChatGPT Stumbles!) — Claude processes ICS calendar files reliably where ChatGPT fails context limits
• How Smart is ChatGPT? — GPT-4 vs GPT-3.5 exam-percentile benchmark comparison
• How To Delete Your Data From ChatGPT — Guide to ChatGPT data deletion rights and privacy controls amid regulatory scrutiny
• MCP: An (Accidentally) Universal Plugin System — MCP as accidentally universal interoperability layer beyond AI context
• Must-Have Prompt Engineering Skills for 2024 — In-demand prompt engineering skills, platforms, and tools
• Navigating the Jagged Technological Frontier | Harvard Business School AI Institute — Field experiment showing AI boosts consultant productivity unevenly across tasks
• New MIT Research Shows Spectacular Increase In White Collar Productivity From ChatGPT — MIT RCT finds ChatGPT makes white-collar workers 37% faster at writing tasks
• Noam Chomsky says A.I. is far from 'true intelligence' and ChatGPT is the 'banality of evil' | Fortune — Chomsky argues LLMs lack true reasoning and exhibit indifference to truth
• One AI Tutor Per Child: Personalized learning is finally here — LLMs enable scalable one-on-one personalized tutoring for every child
• OpenAI CEO Sam Altman says ChatGPT would have passed for an AGI 10 years ago — Sam Altman on shifting AGI definition, hallucinations, and AI training data consent
• OpenAI executives say releasing ChatGPT for public use was a last resort after running into multiple hurdles — and they're shocked by its popularity — OpenAI's surprise at ChatGPT's viral adoption and CEO's AI risk warnings
• OpenAI's 'Pay As You Go' Is the Best Way to Use ChatGPT — OpenAI pay-as-you-go API plan cheaper than ChatGPT Plus subscription
• People Are Using A 'Grandma Exploit' To Break AI - Kotaku — Roleplay persona prompts bypass AI safety guardrails
• Perspectives on the Social Impacts of Reinforcement Learning with Human Feedback — Social and ethical impacts of RLHF across seven societal dimensions
• Practical Guide to Task Automation using ChatGPT & Python — Using ChatGPT and Python APIs to automate common knowledge-work tasks
• Prompt Engineering Embraces New Technique Called Skeleton-Of-Thought — Skeleton-of-thought outlining prompt technique explained
• Prompt Engineering Urges 'Hermeneutic Prompting' As A Powerful Technique Unlocking The True Value Of Generative AI — Hermeneutic circle prompting technique for richer LLM responses
• Re-implementing LangChain in 100 lines of code — Hacker News debate over LangChain's abstraction value
• Samsung workers made a major error by using ChatGPT — Samsung engineers leaked trade secrets via ChatGPT input data retention
• Space Force Gets Scared, Pauses All Use of Generative AI — US Space Force bans generative AI tools on government devices over data security concerns
• State of AI, September 2025: A Broad Primer from an AI Team Lead in Education — Education-focused overview of AI history, capabilities, agents, and AGI trajectory
• Teaching with AI — OpenAI guide for educators using ChatGPT in classrooms
• The False Promise of Imitating Proprietary LLMs — Finetuning open models on ChatGPT outputs mimics style but not factuality or capability
• The Generative AI Copyright Fight Is Just Getting Started — Legal debate over AI training on copyrighted works and fair use
• The Goopification of AI — AI chatbots colonizing the self-help genre via probabilistic text assembly
• The challenges of reinforcement learning from human feedback (RLHF) - TechTalks — RLHF limitations across feedback, reward modeling, and policy
• The founder behind ChatGPT once gave a lecture on how to launch a startup. It's still required viewing 9 years later. — Sam Altman's 2014 Stanford startup lecture key principles still relevant
• Training Your Own LLM using privateGPT — Running a private local LLM on sensitive data without cloud exposure
• Using ChatGPT as a technical writing assistant — ChatGPT as iterative technical writing draft tool
• Vicuna: An Open-Source Chatbot Impressing GPT-4 with 90%* ChatGPT Quality — LLaMA fine-tune on ShareGPT data achieving near-ChatGPT quality for $300
• What Really Made Geoffrey Hinton Into an AI Doomer — Hinton's reasons for leaving Google to warn about accelerating AI risk
• What was 60 Minutes thinking, in that interview with Geoff Hinton? — Gary Marcus rebuttal annotating 60 Minutes Hinton AI interview
• Why some college professors are adopting ChatGPT AI as quickly as students — ChatGPT disruption of higher education from professor perspective
• Wix will let you build an entire website using only AI prompts — Wix AI Site Generator builds full websites from text prompts using ChatGPT and DALL-E