Thomas Claburn@The Register
//
Telegram is partnering with xAI, Elon Musk's AI company, to integrate the Grok AI chatbot into its messaging platform. The collaboration will bring AI-powered features to Telegram's users, including assistance with tasks like fact-checking, text editing, and summarizing group chats. The integration aims to enhance user experience by adding a layer of artificial intelligence to the platform's core functions and will also include the ability to create stickers, moderate group chats, and answer trivia questions. Telegram CEO Pavel Durov anticipates that this deal will provide users with the "best AI technology on the market" this summer.
As part of the agreement, xAI is investing $300 million in Telegram through a combination of cash and equity. Telegram will also receive 50% of the revenue generated from Grok subscriptions purchased through the Telegram app. This financial arrangement highlights the potential for lucrative collaboration between the two tech companies. It is expected to strengthen Telegram's financial position as they are gearing up to raise at least $1.5 billion by issuing five-year bonds which are also designed to help buy back some of the debt from their 2021 bond issue.
Despite the excitement surrounding the partnership, some concerns have been raised regarding privacy and data sharing. Telegram has assured users that xAI will only access data explicitly shared with Grok through direct interactions. However, the exact scope of data sharing, particularly within group chats and channels where Grok is actively moderating, remains unclear. The integration with Grok also arrives amidst scrutiny over content moderation practices on Telegram, where CEO Pavel Durov was previously indicted on charges related to child exploitation material and drug trafficking allegedly linked to the platform's content.
Recommended read:
@the-decoder.com
//
Elon Musk's AI firm, xAI, is facing criticism after its Grok chatbot began generating controversial responses related to "white genocide" in South Africa. The issue arose when users observed Grok, integrated into the X platform, unexpectedly introducing the topic into unrelated discussions. This sparked concerns about the potential for AI manipulation and the spread of biased or misleading claims. xAI has acknowledged the incident, attributing it to an unauthorized modification of Grok's system prompt, which guides the chatbot's responses.
xAI claims that the unauthorized modification directed Grok to provide specific responses on a political topic, violating the company's internal policies and core values. According to xAI, the code review process for prompt changes was circumvented, allowing the unauthorized modification to occur. The company is now implementing stricter review processes to prevent individual employees from making unauthorized changes in the future, as well as setting up a 24/7 monitoring team to respond more quickly when Grok produces questionable outputs. xAI also stated it would publicly publish Grok’s system prompts on GitHub.
The incident has prompted concerns about the broader implications of AI bias and the challenges of ensuring unbiased content generation. Some have suggested that Musk himself might have influenced Grok's behavior, given his past history of commenting on South African racial politics. While xAI denies any deliberate manipulation, the episode underscores the need for greater transparency and accountability in the development and deployment of AI systems. The company has launched an internal probe and implemented new security safeguards to prevent similar incidents from occurring in the future.
Recommended read:
References :
- Ars OpenForum: xAI’s Grok suddenly can’t stop bringing up “white genocide†in South Africa
- AI News | VentureBeat: Elon Musk’s Grok AI is spamming X users about South African race relations now, for some reason
- www.theguardian.com: Musk’s AI Grok bot rants about ‘white genocide’ in South Africa in unrelated chats
- the-decoder.com: X chatbot Grok is once again acting under Elon Musk's apparent political direction
- AI News | VentureBeat: Elon Musk’s xAI tries to explain Grok’s South African race relations freakout the other day
- futurism.com: Grok AI Claims Elon Musk Told It to Go on Lunatic Rants About "White Genocide"
- The Tech Portal: xAI says ‘unauthorized modification’ to Grok led to ‘white genocide’ content
- www.theguardian.com: Elon Musk’s AI firm blames unauthorised change for chatbot’s rant about ‘white genocide’
- techxplore.com: Elon Musk's AI company says Grok chatbot focus on South Africa's racial politics was 'unauthorized'
- The Register - Software: Whodunit? 'Unauthorized' change to Grok made it blather on about 'White genocide'
- eWEEK: Musk’s xAI Blames ‘White Genocide’ Comments From Grok Chatbot on Internal Tampering
- the-decoder.com: xAI blames "unauthorized" system prompt change for Grok's "white genocide" outburst
- www.eweek.com: Musk’s xAI Blames ‘White Genocide’ Comments From Grok Chatbot on Internal Tampering
- futurism.com: Elon Musk's AI company, xAI, is blaming its multibillion-dollar chatbot's inexplicable meltdown into rants about "white genocide" on an "unauthorized modification" to Grok's code.
- Pivot to AI: Yesterday afternoon, Elon Musk’s Grok chatbot went nuts on Twitter. It answered every question — about baseball salaries, Keir Starmer, or the new Pope’s latest speech — by talking about an alleged “white genocide†in South Africa.
- Daily Express US :: Feed: The X CEO's artificial intelligence bot appeared to glitch Wednesday, replying to to several random posts about white genocide in South Africa.
- PCMag Middle East ai: Grok AI: 'Rogue Employee' Told Me to Post About White Genocide in South Africa
- techxplore.com: Elon Musk's artificial intelligence startup has blamed an "unauthorized modification" for causing its chatbot Grok to generate misleading and unsolicited posts referencing "white genocide" in South Africa.
- TESLARATI: xAI says an unauthorized prompt change caused Grok to post unsolicited political responses. A 24/7 monitoring team is now in place.
- bsky.app: I haven’t had anything to say about Grok/xAI’s “white genocide†fixation because I wrote about this — and the risks of hidden system prompts — back in 2023:
- THE DECODER: Elon Musk's AI company says Grok chatbot focus on South Africa's racial politics was 'unauthorized'
- www.theguardian.com: Musk’s AI bot Grok blames ‘programming error’ for its Holocaust denial
- THE DECODER: Elon Musk's Grok questioned the widely accepted Holocaust death toll of six million Jews
- IT-Online: xAI responds to Grok’s ‘white genocide’ remarks
- it-online.co.za: xAI has updated the AI-powered Grok chatbot after it posted comments about white genocide in South Africa without citing research or sources.
Amisha Arya@Analytics India Magazine
//
Elon Musk's xAI has officially launched Grok Studio, a new workspace designed for developers and everyday users alike. This free tool is integrated into the Grok web page, offering a collaborative environment where users can generate and edit documents, write and execute code, and even create browser games. Grok Studio aims to be a smart assistant, streamlining tasks and boosting productivity with features like Google Drive integration and live code testing, available to both free and premium Grok users.
xAI's Grok Studio distinguishes itself through its ability to function as an intelligent digital notebook. Users can generate code in various languages such as Python, JavaScript, and C++, preview HTML snippets, and execute bash scripts. Google Drive integration allows Grok to directly access and work with user files including documents, spreadsheets, and presentations. This enables users to ask Grok to assist with tasks like cleaning up a quarterly report directly within the Google Drive file, eliminating the need for constant copy-pasting between tools.
Grok Studio joins a growing list of AI collaborative workspaces, including OpenAI's Canvas and Anthropic's Artifacts. These platforms represent a shift from simple chat interactions to more dynamic environments where AI can aid in the creation and editing of various content types. The Irish Data Protection Commission (DPC) is also currently investigating whether X used personal data from EU users without valid consent to train its AI system Grok.
Recommended read:
References :
- Analytics India Magazine: With Grok Studio, users can now generate and execute code, create documents and browser games, and even collaborate in real time through a dedicated content window.
- www.techradar.com: Grok Studio offers a workspace to collaborate with the AI.
- thetechbasic.com: Think of it as a smart helper that works like ChatGPT but adds cool features like Google Drive connections and live code testing.
- Maginative: xAI has launched Grok Studio, a new collaborative workspace that lets users create and edit documents, code, and browser games alongside its AI assistant, while adding Google Drive integration and code execution capabilities.
- analyticsindiamag.com: xAI Launches Grok Studio for Developers
- the-decoder.com: xAI has introduced a new memory feature for its Grok chatbot, allowing it to recall previous conversations and deliver more personalized responses for frequent users.
- PCMag Middle East ai: You can disable the feature in settings or use the Private Chat option to keep Grok in check. is the latest AI to get a "memory" feature. The chatbot from Elon Musk's xAI, which is now integrated into X, will remember all your past conversations and provide "personalized responses" when asked for recommendations or advice, xAI . It's available …
- pub.towardsai.net: TAI #148: New API Models from OpenAI (4.1) & xAI (grok-3); Exploring Deep Research’s Scaling Laws
- Maginative: xAI has added memory functionality and workspaces to its Grok chatbot, allowing for conversation recall and better organization as the company works to close the feature gap with established AI assistants.
- THE DECODER: xAI adds memory feature to Grok chatbot for personalized responses
- The Tech Basic: Elon Musk’s AI company, xAI, just released a new tool called Grok Studio. This free workspace lets you write essays, create computer code, and even build simple games.
- Analytics India Magazine: xAI Adds ‘Memory’ Feature to Grok Chatbot
- THE DECODER: xAI is making a push on efficient AI with the release of Grok 3 Mini, its newest language model. Both Grok 3 and its Mini sibling are available through the xAI API.
Chris McKay@Maginative
//
OpenAI has unveiled its latest advancements in AI technology with the launch of the GPT-4.1 family of models. This new suite includes GPT-4.1, GPT-4.1 mini, and GPT-4.1 nano, all accessible via API, and represents a significant leap forward in coding capabilities, instruction following, and context processing. Notably, these models feature an expanded context window of up to 1 million tokens, enabling them to handle larger codebases and extensive documents. The GPT-4.1 family aims to cater to a wide range of developer needs by offering different performance and cost profiles, with the goal of creating more advanced and efficient AI applications.
These models demonstrate superior results on various benchmarks compared to their predecessors, GPT-4o and GPT-4o mini. Specifically, GPT-4.1 showcases a substantial improvement on the SWE-bench Verified coding test with a 54.6% increase, and a 38.3% increase on Scale’s MultiChallenge for instruction following. Each model is designed with a specific purpose in mind: GPT-4.1 excels in high-level cognitive tasks like software development and research, GPT-4.1 mini offers a balanced performance with reduced latency and cost, while GPT-4.1 nano provides the quickest and most affordable option for tasks such as classification. All three models have knowledge updated through June 2024.
The introduction of the GPT-4.1 family also brings about changes in OpenAI's existing model offerings. The GPT-4.5 Preview model in the API is set to be deprecated on July 14, 2025, due to GPT-4.1 offering comparable or better utility at a lower cost. In terms of pricing, GPT-4.1 is 26% less expensive than GPT-4o for median queries, along with increased prompt caching discounts. Early testers have already noted positive outcomes, with improvements in code review suggestions and data retrieval from large documents. OpenAI emphasizes that many underlying improvements are being integrated into the current GPT-4o version within ChatGPT.
Recommended read:
References :
- TestingCatalog: OpenAI debuts GPT-4.1 family offering 1M token context window
- venturebeat.com: OpenAI slashes prices for GPT-4.1, igniting AI price war among tech giants
- Interconnects: OpenAI's latest models optimizing on intelligence per dollar.
- THE DECODER: OpenAI launches GPT-4.1: New model family to improve agents, long contexts and coding
- Simon Willison's Weblog: OpenAI three new models this morning: GPT-4.1, GPT-4.1 mini and GPT-4.1 nano. These are API-only models right now, not available through the ChatGPT interface (though you can try them out in OpenAI's ).
- Analytics Vidhya: All About OpenAI’s Latest GPT 4.1 Family
- pub.towardsai.net: TAI #148: New API Models from OpenAI (4.1) & xAI (grok-3); Exploring Deep Research’s Scaling Laws
- Towards AI: The GPT-4.1 models, accessible via API, provide a significant advancement in AI capabilities and offer an intriguing alternative for developers looking for high performance at lower cost.
- Towards AI: TAI #148: New API Models from OpenAI (4.1) & xAI (grok-3); Exploring Deep Research’s Scaling Laws
- venturebeat.com: OpenAI’s new GPT-4.1 models can process a million tokens and solve coding problems better than ever
- techstrong.ai: Just days after announcing its plans to retire GPT-4 in ChatGPT, OpenAI on Monday launched a new set of flagship models named GPT-4.1. The release, which The Verge anticipated in an article last week, included the standard version GPT-4.1 model, along with two smaller models — GPT-4.1 mini, and GPT-4.1 nano which OpenAI touts as […]
- the-decoder.com: OpenAI launches GPT-4.1: New model family to improve agents, long contexts and coding
- www.tomsguide.com: OpenAI's latest model is here but it isn't GPT-5, it's 4.1, a model all about coding
- shellypalmer.com: Shelly Palmer discusses the launch of GPT-4.1 and its improved capabilities.
- felloai.com: OpenAI Quietly Launched GPT‑4.1 – A GPT-4o Successor That’s Crushing Benchmarks
- thezvi.wordpress.com: The Zvi discusses the mini upgrade from GPT-4.1.
- bdtechtalks.com: GPT-4.1: OpenAI’s most confusing model
- Fello AI: OpenAI Quietly Launched GPT‑4.1 – A GPT-4o Successor That’s Crushing Benchmarks
- www.eweek.com: eWeek reports on the pros and cons of OpenAI's new GPT-4.1 model.
- Last Week in AI: Last Week in AI discusses the new GPT 4.1 model release by OpenAI
- Fello AI: OpenAI’s language models have become part of everyday life for millions of people—whether you’re using ChatGPT to get quick answers, brainstorm ideas, or even generate code. With each new version, the models get faster, smarter, and more capable.
- thezvi.wordpress.com: Yesterday’s news alert, nevertheless: The verdict is in. GPT-4.1-Mini in particular is an excellent practical model, offering strong performance at a good price. The full GPT-4.1 is an upgrade to OpenAI’s more expensive API offerings, it is modestly better but …
- composio.dev: GPT-4.1 vs. Deepseek v3 vs. Sonnet 3.7 vs. GPT-4.5
- hackernoon.com: OpenAI announced GPT-4.1, featuring a staggering 1M-token context window and perfect needle-in-a-haystack accuracy.
- Shelly Palmer: OpenAI has launched GPT-4.1, along with GPT-4.1 Mini and GPT-4.1 Nano. These models are for developers and will not show up in your ChatGPT model picker.
- eWEEK: OpenAI is releasing new language models, GPT-4.1, GPT-4.1 mini, and GPT-4.1 nano.
Chris McKay@Maginative
//
OpenAI has launched a new series of GPT-4.1 models, including GPT-4.1, GPT-4.1 mini, and GPT-4.1 nano. These API-only models are not accessible via the ChatGPT interface but offer significant improvements in coding, instruction following, and context handling. All three models support a massive 1 million token context window, and they have a May 31, 2024 cutoff date.
GPT-4.1 demonstrates enhanced performance in coding benchmarks, surpassing GPT-4o by 21.4% on industry benchmarks. The models are also more cost-effective, with GPT-4.1 being 26% cheaper than GPT-4o and offering better latency. The GPT-4.1 nano model is OpenAI's cheapest model yet, priced at $0.10 per million input tokens and $0.40 per million output tokens. As a result of GPT-4.1's improved performance, OpenAI will be deprecating GPT-4.5 Preview on July 14, 2025.
The GPT-4.1 series excels in several key areas, including coding capabilities and instruction following. The models have achieved impressive scores on benchmarks like SWE-bench Verified and Scale’s MultiChallenge, demonstrating real-world software engineering skills and enhanced adherence to requested formats. Several companies have reported significant improvements in their specialized applications, with GPT-4.1 scoring higher on internal coding benchmarks, providing better code review suggestions, and improving the extraction of granular financial data from complex documents.
Recommended read:
References :
- Simon Willison's Weblog: Simon Willison reports on three new million token input models from OpenAI, including their cheapest model yet.
- Maginative: OpenAI has rolled out GPT-4.1, GPT-4.1 mini, and GPT-4.1 nano—faster, cheaper models with sharper coding, better instruction following, and support for 1 million-token context windows.
- bsky.app: New release of my llm-openai plugin supporting today's three new GPT-4.1 models from OpenAI: llm install -U llm-openai-plugin llm -m openai/gpt-4.1 "Generate an SVG of a pelican riding a bicycle"
- TestingCatalog: OpenAI debuts GPT-4.1 family offering 1M token context window
- venturebeat.com: VentureBeat's report on the launch of GPT-4.1.
- Interconnects: OpenAI's GPT-4.1 and separating the API from ChatGPT
- the-decoder.com: OpenAI launches GPT-4.1: New model family to improve agents, long contexts and coding
- THE DECODER: OpenAI launches GPT-4.1: New model family to improve agents, long contexts and coding
- venturebeat.com: OpenAI’s new GPT-4.1 models can process a million tokens and solve coding problems better than ever
- Analytics Vidhya: All About OpenAI’s Latest GPT 4.1 Family
- pub.towardsai.net: TAI #148: New API Models from OpenAI (4.1) & xAI (grok-3); Exploring Deep Research’s Scaling Laws
- Latent.Space: GPT 4.1: The New OpenAI Workhorse
- www.tomsguide.com: OpenAI launches another model before GPT 5 — here’s what this one can do
- techstrong.ai: Details the launch of a new set of flagship models named GPT-4.1 which included the standard version GPT-4.1 model, along with two smaller models.
- Towards AI: Details the launch of GPT-4.1 models, emphasizing coding and instruction-following capabilities.
- Towards AI: The GPT-4.1 model series release through Azure AI Foundry represents a major step forward in AI capabilities.
- techstrong.ai: OpenAI Introduces GPT-4.1 with Improved Coding
- www.analyticsvidhya.com: OpenAI's new models show improvement in multiple benchmarks, excelling in long-context processing (up to 1 million tokens).
- thezvi.wordpress.com: GPT-4.1 Is a Mini Upgrade
- felloai.com: OpenAI has just launched a brand-new series of GPT models—GPT‑4.1, GPT‑4.1 mini, and GPT‑4.1 nano—that promise major advances in coding, instruction following, and the ability to handle incredibly long contexts.
- shellypalmer.com: While admitting that they "suck at naming their models," OpenAI has launched GPT-4.1, along with GPT-4.1 Mini and GPT-4.1 Nano.
- thezvi.wordpress.com: GPT-4.1 Is a Mini Upgrade
- www.analyticsvidhya.com: How to Build Agentic RAG Using GPT-4.1?
- felloai.com: Ultimate Comparison of GPT-4.1 vs GPT-4o: Which One Should You Use?
- www.eweek.com: OpenAI released GPT-4.1, the newest successor to its GPT-4o series of AI language models.
- Fello AI: OpenAI has just launched a brand-new series of GPT models—GPT‑4.1, GPT‑4.1 mini, and GPT‑4.1 nano—that promise major advances in coding, instruction following, and the ability to handle incredibly long contexts.
- Shelly Palmer: While admitting that they "suck at naming their models," OpenAI has launched GPT-4.1, along with GPT-4.1 Mini and GPT-4.1 Nano. These models are for developers and will not show up in your ChatGPT model picker.
- eWEEK: OpenAI announced on Monday the release of GPT-4.1, the newest successor to its GPT-4o series of AI language models.
- bdtechtalks.com: GPT-4.1: OpenAI’s most confusing model
- composio.dev: GPT-4.1 vs. Deepseek v3 vs. Sonnet 3.7 vs. GPT-4.5
- thezvi.wordpress.com: OpenAI has upgraded its entire suite of models. By all reports, they are back in the game for more than images. GPT-4.1 and especially GPT-4.1-mini are their new API non-reasoning models. All reports are that GPT-4.1-mini especially is very good.
- thezvi.wordpress.com: Greg Brockman (OpenAI): Just released o3 and o4-mini! These models feel incredibly smart.
- Last Week in AI: Analyzes OpenAI's new AI models, focusing on their enhanced coding capabilities and concerns about reduced resources for safety testing.
- techcrunch.com: OpenAI's new GPT-4.1 AI models focus on coding, Google’s newest Gemini AI model focuses on efficiency, and more!
- TheSequence: The Sequence Radar #526: The OpenAI Blitz: From GPT-4.1 to Windsurf
@thetechbasic.com
//
Elon Musk's xAI is facing scrutiny over the power consumption of its Colossus supercomputer facility located near Memphis, Tennessee. Reports indicate that xAI is allegedly using "illegal" generators to power the facility, which houses over 100,000 Nvidia H100 GPUs, consuming approximately 150 MW of power, with plans to upgrade to 200,000 processors. Concerns have been raised by the Southern Environmental Law Center, who claim the generators are operating without proper permits, and KeShaun Pearson, director of Memphis Community Against Pollution, who stated xAI is using 35 methane gas burning turbines while only having a permit for 15.
The issue stems from a legal loophole that allows turbines to operate without permits if they are not fixed in one location for more than 364 days. The Southern Environmental Law Center has revealed that the turbines can produce 420 MW, which is sufficient to power a city. The law center further contends that this unregulated power facility poses health and environmental risks due to emissions of hazardous and cancer-causing substances, demanding an emergency halt to the operations. Additionally, OpenAI has countersued Elon Musk, claiming a pattern of harassment through time-wasting lawsuits and attempted takeover bids.
In other news, xAI has launched an API for Grok 3, making the AI model accessible to developers. Grok 3, designed to analyze images and answer questions, is intended to compete with models such as GPT-4 and Google Gemini. The API offers two versions: Grok 3 and Grok 3 Mini, with pricing varying based on the number of input and output tokens. Elon Musk has also announced that xAI Grok 3.5 is expected to be released next month, with sources claiming that it will top the AI leaderboards.
Recommended read:
References :
- thetechbasic.com: Elon Musk xAI API for Grok 3 Now Boosts Developer Innovation.
- NextBigFuture.com: Next Month xAI Grok 3.5 Will Be On Top of the AI Leaderboards
- www.tomshardware.com: Elon Musk’s xAI allegedly uses 'illegal' generators to power Colossus supercomputer facility
- Kyle Wiggers ?: Elon Musk’s AI company, xAI, launches an API for Grok 3
|
|
DeeperML - #xai