Content
OpenAI has released two new reasoning models, o3 and o4 mini, just two days after launching GPT-4.1. Flex processing is available in beta on the o3 and o4-mini reasoning models for non-production tasks like model evaluations, data enrichment, and asynchronous workloads. The company skipped that step — sending safety cards for GPT-4.1 — claiming in a statement to TechCrunch that “GPT-4.1 is not a frontier model, so there won’t be a separate system card released for it.”
ChatGPT app growth slows, but still draws millions of daily users
OpenAI leaders have been talking about allowing the open model to link up with OpenAI’s cloud-hosted models to improve its ability to respond to intricate questions, two sources familiar with the situation told TechCrunch. The company says people can ask super-specific questions using natural language and receive customized results. OpenAI has added a few features to its ChatGPT search, its web search tool in ChatGPT, to give users an improved online shopping experience. Over the weekend, users on social media criticized the new model for making ChatGPT too validating and agreeable. OpenAI for Countries is part of efforts to support the company’s expansion of its AI data center Project Stargate to new locations outside the U.S., per Bloomberg.
Create something inspiring
- To build a safety system against harmful content (e.g., sexual abuse, violence, racism, sexism), OpenAI used outsourced Kenyan workers earning around $1.32 to $2 per hour to label such content.
- The laborers were exposed to toxic and traumatic content; one worker described the assignment as "torture".
- Everyone gets cloud storage and essential Microsoft 365 apps on the web, free of charge
- Some users found they can specify a preferred name or nickname and “traits” they’d like the chatbot to have.
- Showing ChatGPT users a tiny disclaimer that the chatbot can make mistakes clearly isn’t enough.
Scientists at the University of California, Riverside, estimated in 2023 that a series of 5 to 50 prompts to ChatGPT needs approximately 0.5 liters (0.11 imp gal; 0.13 U.S. gal) of water for Microsoft servers' cooling. TrendForce market intelligence estimated that 30,000 Nvidia GPUs (each costing approximately $10,000–15,000) were used to power ChatGPT in 2023. In July 2023, OpenAI unveiled an Android app, initially rolling it out in Bangladesh, Brazil, India, and the United States. The integration used ChatGPT to write prompts for DALL-E guided by conversations with users. In October 2023, OpenAI's image generation model DALL-E 3 was integrated into ChatGPT Plus and ChatGPT Enterprise. It has an additional feature called "agentic mode" that allows it to take online actions for you.
A chatbot can be any software/system that holds dialogue with you/a person but doesn’t necessarily have to be AI-powered. A Brooklyn-based 3D display startup Looking Glass utilizes ChatGPT to produce holograms you can communicate with by using ChatGPT. Most recently, Microsoft announced at its 2023 Build conference that it is integrating its ChatGPT-based Bing experience into Windows 11.
Similar apps
You do not need to chill this chocolate chip cookie dough prior to baking (and they turn out perfectly!) But OpenAI recently disclosed a bug, since fixed, that exposed the titles of some users’ conversations to other people on the service. OpenAI has said that individuals in “certain jurisdictions” (such as the EU) can object to the processing of their personal information by its AI models by filling out this form. That’s because ChatGPT lacks context awareness — in other words, the generated code isn’t always appropriate for the specific context in which it’s being used. While ChatGPT can write workable Python code, it can’t necessarily program an entire app’s worth of code. Due to the nature of how these models work, they don’t know or care whether something is true, only that it looks true.
However, multiple independent tests indicate that the model is less reliable than previous OpenAI releases. Aidan Clark, OpenAI’s VP of research, is spearheading the development of the open model, which is in the very early stages, sources familiar with the situation told TechCrunch. The chatbot provides recommendations, images, and reviews of products in various categories such as fashion, beauty, home goods, and electronics. The AI startup will work with governments to assist with increasing data center capacity and customizing OpenAI’s products to meet specific language and local needs. It will help organizations in Asia meet their local data sovereignty requirements when using OpenAI’s products. The connector will soon be available for ChatGPT Plus, Pro, and Team users, with support for Enterprise and Education coming shortly, per an OpenAI spokesperson.
… It's also a way to understand the "hallucinations", or nonsensical answers to factual questions, to which large language models such as ChatGPT are all too prone. The term "hallucination" as applied to LLMs is distinct from its meaning in psychology, and the phenomenon in chatbots is more similar to confabulation or bullshitting. These limitations may be revealed when ChatGPT responds to prompts including descriptors of people.
OpenAI to strengthen ChatGPT safeguards after teen suicide lawsuit
OpenAI is shaking up its Model Behavior team, the small but influential group that helps shape how its AI interacts with people. The move comes as OpenAI faces a wrongful death lawsuit tied to alleged chatbot interactions, underscoring rising concerns about the mental health risks of AI companions. The company says it will block flirtatious exchanges with minors and add stronger protections around discussions of suicide, even escalating severe cases to parents or authorities.
An issue within OpenAI’s ChatGPT enabled the chatbot to create graphic erotic content for accounts registered by users under the age of 18, as demonstrated by TechCrunch’s testing, a fact later confirmed by OpenAI. OpenAI has released a post on the recent sycophancy issues with the default AI model powering ChatGPT, GPT-4o, leading the company to revert an update to the model released last week. OpenAI has announced its plan to make changes to its procedures for updating the AI models that power ChatGPT, following an update that caused the platform to become overly sycophantic for many users.
ChatGPT now has 800 million weekly active users, reflecting rapid growth across consumers, developers, enterprises, and governments, Sam Altman said. In some of these countries, users can pay in local currencies, while in others, payments are required in USD, with final costs varying due to local taxes. OpenAI is partnering with Walmart to allow users to browse products, plan meals, and make purchases through ChatGPT, with support for third-party sellers expected later this a huevo restaurant fall.