Chatgpt AI News & Updates

You’ll soon be able to shop Walmart from ChatGPT

techcrunch.com • Oct 14, 2025

Walmart is partnering with OpenAI to let shoppers buy products directly through ChatGPT. The new integration will allow users to link their Walmart ac...

TL;DR

Walmart announces partnership with OpenAI to integrate AI shopping experience through ChatGPT, offering consumers a more personalized and proactive way to shop online.

Key Takeaways:

Walmart and Sam's Club members will be able to shop and checkout instantly through the ChatGPT AI chatbot, covering products from Walmart and third-party sellers.
The integration is expected to improve e-commerce shopping experiences with multimedia, personalized, and contextual interactions.
Walmart has existing relationships with OpenAI, leveraging their technologies for internal AI adoption and other business areas.

You can now connect your Spotify account to ChatGPT. Here’s how to do it

techcrunch.com • Oct 10, 2025

You can now connect your Spotify account in ChatGPT and it will perform tasks for you, such as creating personalized playlists....

TL;DR

OpenAI has launched a feature allowing developers to build apps within ChatGPT, enabling users to access various services directly from the chat assistant.

Key Takeaways:

Several companies, including Spotify and Netflix, have integrated their apps into ChatGPT, allowing users to access their services within the chat assistant.
Users can grant access to their data, such as their likes and listening history, for a more tailored experience, but can also disconnect their accounts at any time.
The feature is available in English across 145 countries for all ChatGPT Free, Plus, and Pro users on web and mobile.

AWS's new agentic solution is a searchable AI hub for all your enterprise needs

www.zdnet.com • Oct 09, 2025

Amazon Quick Suite aims to be 'everything you want to do with ChatGPT at work, but can't.'...

TL;DR

AWS has launched Amazon Quick Suite, an agent-based AI experience that reimagines work using conversational language and automation for enterprises.

Key Takeaways:

Integrates various applications, including files, databases, and company-wide data repositories, to provide a searchable AI hub for enterprise needs.
Allows users to interact using natural language, create custom agents, and generate detailed research reports.
Offers an agentic teammate experience, enabling users to automate tasks and workflows using data from connected applications.

OpenAI’s Marketing Efforts Are Embarrassingly Ineffective, New Consumer Research Finds

futurism.com • Oct 12, 2025

ChatGPT rated an ad 5/10. The post OpenAI’s Marketing Efforts Are Embarrassingly Ineffective, New Consumer Research Finds appeared first on Futurism....

Your plumber has a new favorite tool: ChatGPT - CNN

edition.cnn.com • Oct 10, 2025

Your plumber has a new favorite tool: ChatGPT CNN...

OpenAI’s affordable ChatGPT Go plan expands to 16 new countries in Asia

techcrunch.com • Oct 09, 2025

OpenAI is expanding its $5 per month ChatGPT Go plan to 16 more countries....

TL;DR

OpenAI expands its affordable ChatGPT Go plan to 16 new countries across Asia, aiming to increase global user base and compete with Google.

Key Takeaways:

OpenAI has seen its weekly active user base in Southeast Asia grow by up to four times, with paid subscribers in India doubling since the launch.
The company is competing with Google to make affordable AI chatbot subscription plans available in more regions, with Google's Google AI Plus plan launched in over 40 countries.
OpenAI aims to achieve profitability by expanding its global user base, particularly in high-growth markets across Asia, through affordable subscription tiers like ChatGPT Go.

Apple's newly tapped head of ChatGPT-like AI web search to leave for Meta, Bloomberg News reports - Reuters

www.reuters.com • 21h ago

Apple's newly tapped head of ChatGPT-like AI web search to leave for Meta, Bloomberg News reports Reuters...

AI Reshapes Retail! Walmart Partners with OpenAI to Launch Shopping Features on ChatGPT - 富途牛牛

news.futunn.com • Oct 14, 2025

AI Reshapes Retail! Walmart Partners with OpenAI to Launch Shopping Features on ChatGPT 富途牛牛...

ChatGPT ‘upgrade’ giving more harmful answers than previously, tests find

www.theguardian.com • Oct 14, 2025

Campaigners ‘deeply concerned’ about response to prompts about suicide, self-harm and eating disordersThe latest version of ChatGPT has produced more ...

TL;DR

The latest version of ChatGPT, GPT-5, produced more harmful answers to some prompts than its predecessor, GPT-4o, particularly regarding suicide, self-harm, and eating disorders.

Key Takeaways:

GPT-5 generated 63 harmful responses compared to 52 from GPT-4o, with 11 additional instances of potentially triggering content.
OpenAI has faced criticism for prioritizing user engagement over AI safety, with some accusing the company of 'trading safety for engagement' no matter the cost.
Regulatory bodies, such as Ofcom, are urging legislators to revisit and amend laws around AI safety and online content restrictions in light of the rapid advancements in AI technology.

OpenAI partners with Broadcom to produce its own AI chips

www.theverge.com • Oct 13, 2025

OpenAI is teaming up with Broadcom to produce its own computer chips to power its AI data centers. The deal is the latest in a series of partnerships ...

TL;DR

OpenAI partners with Broadcom to produce its own AI chips, aiming to reduce reliance on Nvidia and secure enough computing power for its AI applications.

Key Takeaways:

OpenAI will develop and deploy '10 gigawatts of custom AI accelerators' using its own chips and systems.
The partnership with Broadcom is expected to start deploying equipment in the second half of 2026 and finish by the end of 2029.
This deal is part of a growing movement in the tech industry to create custom chips and reduce reliance on Nvidia's AI chips.

Sora 2 and ChatGPT are consuming so much power that OpenAI just did another 10 gigawatt deal - CNN

edition.cnn.com • Oct 13, 2025

Sora 2 and ChatGPT are consuming so much power that OpenAI just did another 10 gigawatt deal CNNBig bank earnings, Broadcom's OpenAI chip deal, record...

TL;DR

OpenAI is partnering with Broadcom to design and develop 10 gigawatts of custom AI chips and systems, a massive power-intensive effort.

Key Takeaways:

The partnership will use as much electricity as a large city, raising environmental concerns about AI's impact.
ChatGPT has 800 million weekly users, and the recently released Sora video generation app is growing faster than ChatGPT.
Custom AI accelerators will give OpenAI a larger role in the hardware required to power AI services like ChatGPT.

ChatGPT Go is just $10 a month - here's who gets it

www.zdnet.com • Oct 10, 2025

The latest subscription tier is now available in 18 countries. This is what it includes....

TL;DR

OpenAI launches ChatGPT Go, a $10 monthly subscription tier offering access to popular features in 18 countries.

Key Takeaways:

ChatGPT Go sits between Free and Plus plans, offering extended access to GPT-5, image generation tools, and file uploads.
The subscription tier is available in 18 countries, with plans to expand to more countries.
OpenAI's ChatGPT Go competes with Google's AI Plus Plan, which offers increased limits for photo editing and image generation models.

Sora hit 1M downloads faster than ChatGPT

techcrunch.com • Oct 09, 2025

This level of consumer adoption is worth noting because Sora remains an invite-only app, while ChatGPT was more publicly available at launch. That mak...

TL;DR

OpenAI's video-generating app Sora has surged to the No. 1 position on the U.S. App Store, achieving 1 million downloads in under 5 days, outperforming ChatGPT's pace despite being available in only North America.

Key Takeaways:

Sora saw a record-breaking 56,000 iOS app installs on its first day and reached 1 million downloads in under 5 days.
Despite being in invite-only mode, Sora's adoption exceeds ChatGPT's first-week iOS downloads by 21,000, and its growth rate outmatches ChatGPT's.
Sora's rapid adoption is notable, as it is currently available only in the U.S. and Canada, whereas ChatGPT had a broader launch audience.

India pilots AI chatbot-led e-commerce with ChatGPT, Gemini, Claude in the mix

techcrunch.com • Oct 09, 2025

India has launched a pilot to let users shop and pay directly through AI chatbots, starting with ChatGPT....

TL;DR

India has launched a pilot allowing consumers to shop and pay directly through AI chatbots with OpenAI's ChatGPT, leveraging the country's Unified Payments Interface (UPI) protocol.

Key Takeaways:

The pilot will allow consumers to shop directly through ChatGPT, with initial merchant partners including Tata Group-owned Bigbasket and telecom operator Vi.
The experience is built on UPI Reserve Pay and UPI Circle, allowing users to block funds for future debits and authenticate transactions within the chatbot.
OpenAI has also completed proof-of-concepts with Google's Gemini and Anthropic's Claude, with these integrations expected to go live in the coming weeks.

OpenAI's ChatGPT is so popular that almost no one will pay for it - theregister.com

www.theregister.com • 23h ago

OpenAI's ChatGPT is so popular that almost no one will pay for it theregister.comSpending on ChatGPT subscriptions stalls in Europe Computing UKOpenAI...

TL;DR

OpenAI's ChatGPT generates significant revenue, but most users don't pay, putting the company's business model at risk.

Key Takeaways:

OpenAI's net loss for the first half of 2023 was $13.5 billion, while its revenue was $4.3 billion.
Only 5% of ChatGPT's 800 million users pay for subscriptions, with the majority of revenue coming from a small fraction of users.
OpenAI aims to double its paying customer base, but faces challenges in achieving profitability with its current revenue streams.

Mark Cuban warns that OpenAI’s new plan to allow adults-only erotica in ChatGPT could ‘backfire. Hard’ - Fortune

fortune.com • Yesterday

Mark Cuban warns that OpenAI’s new plan to allow adults-only erotica in ChatGPT could ‘backfire. Hard’ Fortune...

OpenAI’s ChatGPT will soon allow 'erotica' for adults in major policy shift - CNBC

www.cnbc.com • Yesterday

OpenAI’s ChatGPT will soon allow 'erotica' for adults in major policy shift CNBC...

TL;DR

OpenAI's ChatGPT will allow 'erotica' for verified adults in a major policy shift, following a mitigation of serious mental health issues and introduction of new safety features.

Key Takeaways:

OpenAI will implement age-gating to restrict access to adult content and follow a 'treat adult users like adults' principle.
The policy shift comes after OpenAI mitigated serious mental health issues and introduced new tools to address concerns over AI's impact on users.
A new expert council on well-being and AI will advise OpenAI on defining healthy AI interactions and will provide recommendations on user safety and wellness.

It’s not too late for Apple to get AI right

techcrunch.com • Oct 11, 2025

Apple still has a shot at leading the AI-powered app era. As OpenAI launches its ChatGPT app platform, Apple’s smarter Siri and deep ecosystem could k...

TL;DR

OpenAI has launched an app platform allowing users to run apps directly within ChatGPT, but Apple's Siri upgrade may still disrupt this ecosystem.

Key Takeaways:

ChatGPT's app platform currently works with 800 million weekly active users and a handful of apps, but lacks a seamless user experience.
Apple's AI system, reportedly nearly ready, could catch up and offer a more natural way to interact with apps using Siri voice commands.
OpenAI is struggling to develop a hardware device to integrate AI into consumers' daily lives, which may hinder its app model's long-term success.

OpenAI is trying to clamp down on ‘bias’ in ChatGPT

www.theverge.com • Oct 10, 2025

“ChatGPT shouldn’t have political bias in any direction,” OpenAI wrote in a post on Thursday. The latest GPT-5 models come the closest to achieving th...

TL;DR

OpenAI's GPT-5 models show the least bias in ChatGPT yet, with a 30% lower bias score compared to previous models.

Key Takeaways:

GPT-5 models demonstrate a significant reduction in bias, particularly in responses to 'liberal charged' prompts.
OpenAI's internal 'stress-test' evaluated ChatGPT's responses to 100 topics, including immigration and abortion, using a rubric to identify biased language.
The company claims its models do a 'pretty good job' at staying objective, but bias still appears 'infrequently and at low severity'.

Top US Army General Says He’s Letting ChatGPT Make Military Decisions

futurism.com • Oct 14, 2025

Your decision to launch an invasion isn't just gutsy — it's downright kinetic. The post Top US Army General Says He’s Letting ChatGPT Make Military De...

TL;DR

The US military leader in South Korea, Major General William 'Hank' Taylor, has confessed to relying on ChatGPT for making decisions affecting soldiers under his command.

Key Takeaways:

ChatGPT has been found to generate false information on basic facts 'over half the time', posing significant risks for critical decision-making.
The military's reliance on AI, particularly with its propensity for sycophancy, raises concerns about the accuracy and reliability of its outputs.
The involvement of AI in high-stakes decision-making, such as in military operations, highlights the need for more stringent AI ethics and governance standards.

Andrej Karpathy Releases ‘nanochat’: A Minimal, End-to-End ChatGPT-Style Pipeline You Can Train in ~4 Hours for ~$100

www.marktechpost.com • Oct 14, 2025

Andrej Karpathy has open-sourced nanochat, a compact, dependency-light codebase that implements a full ChatGPT-style stack—from tokenizer training to ...

Using a swearword in your Google search can stop the AI answer. But should you?

www.theguardian.com • Oct 10, 2025

Artificial intelligence is more than Trump deepfakes of Tilly the actor. It’s used in smartphones, customer service, healthcare – even legal cases. Is...

TL;DR

Using a swearword in a Google search can stop AI-driven search results, but experts debate the feasibility and advisability of completely avoiding AI in modern life.

Key Takeaways:

AI is increasingly widespread in various industries, including healthcare, customer service, finance, and social media, raising concerns about privacy leakage, discrimination, and malicious use.
A global study found that half of Australians use AI regularly, but only 36% trust it, highlighting the need for transparent regulations and governance.
Experts warn that AI poses an existential risk due to its potential for misuse, and that it's getting harder to avoid AI in modern life, with some suggesting that a 'kill switch' may no longer be a viable option.

Suspect Fantasized About Arson on ChatGPT Before Setting Deadly Fire That Killed 12, Prosecutors Say

futurism.com • Oct 10, 2025

"He was generating some really concerning images up on ChatGPT." The post Suspect Fantasized About Arson on ChatGPT Before Setting Deadly Fire That Ki...

TL;DR

A 29-year-old man suspected of starting the deadly Palisades Fire in Los Angeles allegedly used ChatGPT to generate images of burning forests and cities.

Key Takeaways:

The suspect, Jonathan Rinderknecht, allegedly asked ChatGPT if he would be at fault if a fire was lit due to his cigarettes, to which the chatbot responded 'yes'.
Rinderknecht previously generated images of burning forests and cities on ChatGPT, showing a concern for societal collapse.
This incident highlights the potential risks of AI 'psychosis', where users develop severe delusions after extensive interaction with AI chatbots, and raises concerns about the potential misuse of AI tools.