ChatGPT
- OSINT with ChatGPT Search Enginefosstodon.org Dimitris Vagiakakos 🇪🇺🇬🇷 (@sv1sjp@fosstodon.org)
Attached: 1 image #OSINT with the new #ChatGPT Search Engine . The goal was to find as much information as possible about myself, my family members, and partners, using only my #HamRadio callsign (SV1SJP) as the starting point. ChatGPT Search Engine seems promising :)) https://sv1sjp.github.io/ar...
- Reasoning failures highlighted by Apple research on LLMsappleinsider.com Reasoning failures highlighted by Apple research on LLMs
A new paper from Apple's artificial intelligence scientists has found that engines based on large language models, such as those from Meta and OpenAI, still lack basic reasoning skills.
cross-posted from: https://programming.dev/post/20491311 >
- All the chats for the last month -- gone
Free account of ChatGPT
Yesterday, all All the chats for the last or even 2 months -- no more.
No notification before hand, no notification whether it's bug or they want me to upgrade. Nothing. I've cleared the cookies and local storage in the browser, re-logged in. To no avail.
What the fucking wtf-king fuck?
- ChatGPT voice mode with a toddler
I'm pretty impressed with how well it's able to understand him, and how quickly it's able to respond, especially with two people talking, interrupting, changing languages, etc.
- AI Is Evolving Faster Than Experts, Including Bill Gates, Imaginedwww.cnet.com AI Is Evolving Faster Than Experts, Including Bill Gates, Imagined
Get up to speed on the rapidly evolving world of AI with our roundup of the week's developments.
- Is chatGPT currently down or is it just me?
The is it just me says it's up but what I get is the input box but no other design elements show up. I can type in the input box but I can't submit it. It has been like this since 9/20 in the morning. I have cleared my browser cache.
- Google NotebookLM allows you to automatically generate a two-person podcast based on documents or links. It simulates emotions while talking.
NotebookLM is a multi-purpose notebook with AI features. You can create notebooks, feed content, ask questions about the content, save individual answers, and much more. Additionally, it can create a podcast based on your content.
- When having a meal with classmates or colleagues, someone takes photos of everyone and shares it in a group chat without notifying beforehand. How can I politely decline
When having a meal with classmates or colleagues, someone takes photos of everyone and shares it in a group chat without notifying beforehand. How can I politely decline because I don't want to be photographed? @ChatGPT@lemmings.world
- I was unhappy with past therapists and turned to ChatGPT. It helped me open up about trauma and make changes in my life.www.businessinsider.com I was unhappy with past therapists and turned to ChatGPT. It helped me open up about trauma and make changes in my life.
I made a customized ChatGPT therapist that aligned with my values. I found I could ask it questions I usually felt guilty talking to humans about.
- Sam Altman indicated it's impossible to create ChatGPT without copyrighted material, but a new study claims 57% of the content on the internet is AI-generated and is subtly killing quality search resuwww.windowscentral.com Sam Altman indicated it's impossible to create ChatGPT without copyrighted material, but a new study claims 57% of the content on the internet is AI-generated and is subtly killing quality search results
The knowledge scope of AI tools is getting thinner as AI-generated content floods the internet, limiting training material.
- OpenAI starts rolling out its Her-like voice mode for ChatGPT - The Vergewww.theverge.com OpenAI starts rolling out its Her-like voice mode for ChatGPT
The Scarlett Johansson-like voice isn’t part of the product.
- I made a question to Llama 3.1 8B, but forgot to paste the context text
Notes:
- I hit two times the <ENTER> key after the question. Don’t know if that affected the answer.
- I don't have any preloaded context.
- GPT-4o new voice mode rolls out this week!
Screenshot taken last week. Will roll out this week (week of July 28).
- “Copyright traps” could tell writers if an AI has scraped their workwww.technologyreview.com “Copyright traps” could tell writers if an AI has scraped their work
The technique has been used throughout history, but now could be a tool in one of the biggest fights in artificial intelligence.
- Llama 3.1 (already available in ollama) was release with many improvements since Llama 3ai.meta.com Introducing Llama 3.1: Our most capable models to date
Bringing open intelligence to all, our latest models expand context length, add support across eight languages, and include Meta Llama 3.1 405B— the first frontier-level open source AI model.
- Create Better AI Images With These Expert Prompt Writing Tipswww.cnet.com Create Better AI Images With These Expert Prompt Writing Tips
A descriptive prompt is key to bringing your artistic vision to life. These are my expert tips for writing the best AI image prompts.
- PSA: ChatGPT can be used anonymously.
ChatGPT can be used without logging in. But there is a catch, we can't opt out of data training.
Never heard about it anywhere. OpenAI may have secretly released it.
- Hanging onto Windows 10 to avoid Copilot AI? I’ve got bad news for youwww.techradar.com Hanging onto Windows 10 to avoid Copilot AI? I’ve got bad news for you
New Copilot app experience is coming to both Windows 10 and Windows 11 simultaneously, presumably to help stoke usage.
- The Words That Give Away Generative AI Textwww.wired.com The Words That Give Away Generative AI Text
From “delves” to “showcasing,” certain words boomed in usage after LLMs became mainstream.
- Tokens are a big reason today's generative AI falls short | TechCrunchtechcrunch.com Tokens are a big reason today's generative AI falls short | TechCrunch
Tokenization, the process by which many generative AI models make sense of data, is flawed in key ways.
- Cant use ChatGPT without an account anymore?
There is no longer an option to use ChatGPT without an ID on chatgpt.com for me. Is anyone else having the same problem?
- Codeintegrity-ai/mutahunter: Open Source, Language Agnostic LLM-based mutation testing toolgithub.com GitHub - codeintegrity-ai/mutahunter: Open Source Language Agnostic LLM-based mutation testing tool
Open Source Language Agnostic LLM-based mutation testing tool - codeintegrity-ai/mutahunter
Check out our open-source, language-agnostic mutation testing tool using LLM agents here: https://github.com/codeintegrity-ai/mutahunter
Mutation testing is a way to verify the effectiveness of your test cases. It involves creating small changes, or “mutants,” in the code and checking if the test cases can catch these changes. Unlike line coverage, which only tells you how much of the code has been executed, mutation testing tells you how well it’s been tested. We all know line coverage is BS.
That’s where Mutahunter comes in. We leverage LLM models to inject context-aware faults into your codebase. As the first AI-based mutation testing tool, Our AI-driven approach provides a full contextual understanding of the entire codebase by using the AST, enabling it to identify and inject mutations that closely resemble real vulnerabilities. This ensures comprehensive and effective testing, significantly enhancing software security and quality. We also make use of LiteLLM, so we support all major self-hosted LLM models
We’ve added examples for JavaScript, Python, and Go (see /examples). It can theoretically work with any programming language that provides a coverage report in Cobertura XML format (more supported soon) and has a language grammar available in TreeSitter.
Here’s a YouTube video with an in-depth explanation: https://www.youtube.com/watch?v=8h4zpeK6LOA
Here’s our blog with more details: https://medium.com/codeintegrity-engineering/transforming-qa-mutahunter-and-the-power-of-llm-enhanced-mutation-testing-18c1ea19add8
Check it out and let us know what you think! We’re excited to get feedback from the community and help developers everywhere improve their code quality.
- What happened to GPT -4o Censorship This Weekend?
Over the weekend (this past Saturday specifically), GPT-4o seems to have gone from capable and rather free for generating creative writing to not being able to generate basically anything due to alleged content policy violations. It'll just say "can't assist with that" or "can't continue." But 80% of the time, if you regenerate the response, it'll happily continue on its way.
It's like someone updated some policy configuration over the weekend and accidentally put an extra 0 in a field for censorship.
GPT-4 and GPT 3.5 seem unaffected by this, which makes it even weirder. Switching to GPT 4 will have none of the issues that 4o is having.
I noticed this happening literally in the middle of generating text.
See also: https://old.reddit.com/r/ChatGPT/comments/1droujl/ladies_gentlemen_this_is_how_annoying_kiddie/
https://old.reddit.com/r/ChatGPT/comments/1dr3axv/anyone_elses_ai_refusing_to_do_literally_anything/
- ChatGPT would have been so much useful and trustworthy if it is able to accept that it doesn't know an answer.
Small rant : Basically, the title. Instead of answering every question, if it instead said it doesn't know the answer, it would have been trustworthy.
- Ilya Sutskever (co-founder of OpenAI) created a new company called Safe Superintelligence Inc. (SSI)www.theverge.com OpenAI’s former chief scientist is starting a new AI company
The startup’s goal is to build a safe and powerful AI system.
Company website: https://ssi.inc
- Apple integrates ChatGPT into Siri, iOS, and Mac OSarstechnica.com Apple integrates ChatGPT into Siri, iOS, and Mac OS
ChatGPT is treated like a search engine for Siri, and generates text and images for apps. See full article...
