Leo Rodman August 11, 2025
GPT-5 just launched and ChatGPT users are already unhappy with OpenAI’s next gen chatbot
Is 5 really better than 4?
www.techradar.com/ai-platforms-assistants/chatgpt/chatgpt-users-are-not-happy-with-gpt-5-launch-as-thousands-take-to-reddit-claiming-the-new-upgrade-is-horrible
Introducing GPT-5 - Resource | OpenAI Academy
GPT-5 is the newest OpenAI experience that powers ChatGPT with smarter reasoning, more reliable answers, and deeper insights.
academy.openai.com/public/resources/intro-gpt-5
GPT-5 and the new era of work
GPT-5 is OpenAI’s most advanced model—transforming enterprise AI, automation, and workforce productivity in the new era of intelligent work.
openai.com/index/gpt-5-new-era-of-work
Here’s everything in GPT-5 that’s new and different from OpenAI’s previous AI models
If you’re wondering what all the hoopla is about, and what makes GPT-5 better than, or different from, the company’s previous AI models, here’s a quick rundown of some of the key new features and functionality.
fortune.com/2025/08/07/gpt-5-everything-new-different-hallucinations-personalities-vibecoding-agents-openai
OpenAI priced GPT-5 so low, it may spark a price war | TechCrunch
Developers who have had early access are touting GPT-5's pricing as much better than competing models.
techcrunch.com/2025/08/08/openai-priced-gpt-5-so-low-it-may-spark-a-price-war
Forcing LLMs to be evil during training can make them nicer in the long run
New Anthropic research shows that undesirable LLM traits can be detected—and even prevented—by examining and manipulating the model’s inner workings.
www.technologyreview.com/2025/08/01/1120924/forcing-llms-to-be-evil-during-training-can-make-them-nicer-in-the-long-run
AI site Perplexity uses “stealth tactics” to flout no-crawl edicts, Cloudflare says
The allegations are the latest to accuse Perplexity of improper web crawling.
arstechnica.com/information-technology/2025/08/ai-site-perplexity-uses-stealth-tactics-to-flout-no-crawl-edicts-cloudflare-says
ChatGPT adds mental health guardrails after reports of bot feeding people’s delusions
ChatGPT users who converse with the bot for an extended amount of time will now receive reminders encouraging them to take a break
www.the-independent.com/tech/chatgpt-openai-mental-health-psychosis-b2801927.html
Apple is working on a ChatGPT alternative — what we know about the 'Answers Engine'
A push from Apple to win the AI game
www.tomsguide.com/ai/apple-is-working-on-a-chatgpt-alternative-but-is-it-too-late
US executive branch agencies will use ChatGPT Enterprise for just $1 per agency
Questions linger about ideological bias in models as well as data security.
arstechnica.com/ai/2025/08/openai-announces-deal-to-offer-chatgpt-to-us-executive-branch-at-almost-no-cost
xAI's legal chief steps down after whirlwind year | TechCrunch
Despite calling his time at xAI "incredible" and working with Musk "the adventure of a lifetime," Robert Keele said he couldn't keep "riding two horses at once — the family and the job."
techcrunch.com/2025/08/07/xais-legal-chief-steps-down-after-whirlwind-year
Truth Social's AI search is powered by Perplexity, but the platform can set limits on sources | TechCrunch
The search engine, dubbed Truth Search AI, is already available on the web version of Truth Social, with public Beta testing on the iOS and Android apps planned for “the near future.”
techcrunch.com/2025/08/07/truth-socials-ai-search-is-powered-by-perplexity-but-the-platform-can-set-limits-on-sources
Google search boss says AI isn’t killing search clicks
Liz Reid says Google’s data shows AI is generating consistent clicks and better experiences.
arstechnica.com/google/2025/08/google-search-boss-says-ai-isnt-killing-search-clicks
The backlash against Duolingo going 'AI-first' didn't even matter | TechCrunch
Duolingo faced significant backlash for announcing that it would become an "AI-first" company -- and yet, the app posted strong financials this quarter.
techcrunch.com/2025/08/07/the-backlash-against-duolingo-going-ai-first-didnt-even-matter
Some AI tools don’t understand biology yet
A collection of new studies on gene activity shows that AI tools aren’t very good.
arstechnica.com/science/2025/08/some-ai-tools-dont-understand-biology-yet
Reply