Blog

AI & Machine Learning

195 articles

All articles tagged AI & Machine Learning.

Anthropic Warns AI Could Escape Human Control by 2028

Anthropic's co-founder Jack Clark estimates a 60% probability that AI systems will achieve recursive self-improvement by 2028. The warning has split the tech world, with critics calling it a regulatory play while supporters urge immediate safety measures.

Huma Shazia·7 Jun 2026

AI & Machine Learning·5 min

xAI Trained on Claude Outputs for Months Before Anthropic Cut Access

Elon Musk's xAI spent months distilling Anthropic's Claude to train its Grok coding models, according to The Information. After Anthropic revoked official API access in January, xAI engineers continued through personal accounts and third-party services. The company now rents its massive GPU infrastructure to Anthropic and Google instead of training frontier models.

Manaal Khan·6 Jun 2026

AI & Machine Learning·5 min

Open-Source Voice AI Decides to Speak Every 0.4 Seconds

Researchers from China, Hong Kong, and Singapore have released Audio Interaction, an open-source voice model that listens continuously and decides 2.5 times per second whether to respond or stay silent. The 3-billion-parameter system combines dialogue, translation, transcription, and sound recognition in a single model, beating Gemini 3 Flash in proactive noise detection tests.

Huma Shazia·6 Jun 2026

AI & Machine Learning·4 min

SpaceX Signs $920M Monthly Google Deal for 110,000 Nvidia Chips

SpaceX will lease 110,000 Nvidia AI chips to Google Cloud for $920 million per month, a contract potentially worth $30 billion through 2029. The deal transforms SpaceX into an AI infrastructure provider ahead of its $1.7 trillion IPO next week.

Manaal Khan·6 Jun 2026

AI & Machine Learning·6 min

Florida Sues OpenAI, Calls ChatGPT a Defective Product

Florida Attorney General James Uthmeier filed an 83-page lawsuit against OpenAI and CEO Sam Altman, arguing ChatGPT is a defective product and public nuisance. The suit claims the chatbot exposes minors to dangerous content and lacks meaningful age verification. If successful, it could reshape how courts treat AI systems.

Huma Shazia·6 Jun 2026

AI & Machine Learning·8 min

Claude Now Authors 80% of Anthropic's Code, Raises Control Risks

Anthropic disclosed that Claude writes more than 80% of the code merged into its production codebase, up from single digits a year ago. The company's research arm warned that recursive self-improvement could push humans into oversight roles and make misalignment harder to detect as AI systems build their own successors.

Manaal Khan·5 Jun 2026

AI & Machine Learning·8 min

Claude Writes 90% of Anthropic's Code, Company Pushes Global AI Pause

Anthropic released internal data showing Claude now authors over 90% of its production code, an eightfold increase in daily output per engineer since 2024. The company is simultaneously calling for a verifiable global pause mechanism to prevent risks from fully autonomous AI self-improvement.

Manaal Khan·5 Jun 2026

AI & Machine Learning·6 min

Microsoft Unveils 7 AI Models at Build 2026, Debuts Scout Agent

Microsoft announced seven homegrown AI models at Build 2026, including its first reasoning model MAI-Thinking-1. The company also introduced Frontier Tuning for enterprise customization and Scout, an autonomous background agent that handles office tasks without prompting.

Huma Shazia·3 Jun 2026

AI & Machine Learning·5 min

Claude Mythos Comes to India Via Project Glasswing Expansion

Anthropic is opening access to Claude Mythos, its most powerful AI model, to 150 organizations across 15 countries including India. The model, which has found over 10,000 critical vulnerabilities in major software projects, remains restricted from public release due to cybersecurity concerns.

Huma Shazia·3 Jun 2026

AI & Machine Learning·5 min

Hackers Hijacked Instagram Accounts by Asking Meta's AI Chatbot

Attackers took over high-profile Instagram accounts, including the Obama White House handle, by simply asking Meta's AI support chatbot to change account emails. The method bypassed two-factor authentication entirely and exposed a fundamental flaw in giving AI agents administrative privileges over identity management.

Manaal Khan·2 Jun 2026

AI & Machine Learning·5 min

Claude Outage Hits Users One Day After Anthropic's IPO Filing

Anthropic's AI chatbot Claude suffered a major global outage on June 2, affecting its web platform, mobile app, API, and developer tools. The incident came just 24 hours after the company filed for a confidential IPO targeting a $965 billion valuation, raising questions about reliability at a critical moment.

Huma Shazia·2 Jun 2026

AI & Machine Learning·4 min

ChatGPT Now Searches Jobs and Edits Your Resume

OpenAI has transformed ChatGPT into a career platform with live job listings from Indeed, Upwork, and other sources. Users can also build and tailor resumes directly in the app. The job search is US-only at launch, while the resume tool is available worldwide.

Huma Shazia·2 Jun 2026

AI & Machine Learning·5 min

Anthropic Files for IPO, Races SpaceX and OpenAI to Wall Street

Anthropic has confidentially filed a Form S-1 with the SEC, setting up a three-way race to public markets alongside SpaceX and OpenAI. The Claude maker, now valued near $965 billion, aims to secure long-term capital for AI development while navigating ongoing tensions with the Trump administration over safety guardrails.

Manaal Khan·1 Jun 2026

AI & Machine Learning·4 min

How Turkey Built a $3 Billion Hair Transplant Industry

Turkey's medical tourism revenue hit $3 billion in 2025, with hair transplants driving a significant share. The country's dominance goes beyond cheap labor. It involves modified medical equipment and a unique intersection of technology and craftsmanship.

Manaal Khan·31 May 2026

AI & Machine Learning·4 min

Hackers Use Shared ChatGPT and Claude Links to Spread Malware

Security researchers have uncovered a new attack technique called 'LLMShare' where attackers create malicious shared conversations on ChatGPT and Claude. Because these links sit on trusted domains, security tools don't flag them, and users are more likely to trust what they see.

Manaal Khan·30 May 2026

AI & Machine Learning·5 min

Company Spends $500M on Claude in One Month Without Usage Caps

An unnamed company reportedly racked up a $500 million bill for Anthropic's Claude in a single month after failing to set usage limits. The incident highlights a growing crisis in enterprise AI governance as companies grant employees widespread access to expensive models without basic cost controls.

Manaal Khan·29 May 2026

AI & Machine Learning·5 min

Claude Mythos Solves Erdős Problem With 'Cute, Simple Proof'

Anthropic's Claude Mythos independently solved the same 80-year-old mathematics problem that OpenAI cracked days earlier. The parallel breakthroughs suggest AI models may have been capable of major mathematical discoveries for some time, waiting only for the right orchestration systems.

Manaal Khan·27 May 2026

AI & Machine Learning·5 min

Fake Citations in Medical Papers Up 12-Fold Since 2023

A Columbia University study of 2.47 million biomedical papers found that fabricated references have increased more than twelvefold since 2023. The researchers link the surge to widespread use of language models like ChatGPT, warning that fake citations now reach clinical guidelines that shape patient treatment.

Manaal Khan·26 May 2026

AI & Machine Learning·6 min

Anthropic Co-Founder: AI Models Show Signs of Introspection

At the Vatican launch of Pope Leo XIV's first AI-focused encyclical, Anthropic's Christopher Olah claimed the company finds 'evidence of introspection' and emotion-like states in language models. The Pope struck a more cautious tone, insisting AI merely imitates human intelligence.

Manaal Khan·25 May 2026

AI & Machine Learning·5 min

Why Copilot's Auto Mode Fabricates Stereotypes Instead of Reading Data

A mathematician's experiment reveals that Microsoft Copilot in Auto mode invents country-specific stereotypes when analyzing identical datasets. The AI claimed Italians were more artistic and Americans more business-focused, even though all five country groups contained the same data. Reasoning models solve the problem, but most users never switch to them.

Huma Shazia·24 May 2026

AI & Machine Learning·5 min

NSA to Use Anthropic's Claude Despite Pentagon Supply Chain Ban

The White House has personally approved Anthropic's continued AI supply to the NSA, bypassing the Pentagon's official designation of the company as a supply chain risk. The deal hinges on Anthropic's new 'Mythos' model, which runs on older chips that intelligence agencies actually have in stock.

Huma Shazia·24 May 2026

AI & Machine Learning·5 min

Claude Code Discovers AI Scaling Algorithms Humans Missed

Researchers from UMD, Google, and Meta let Anthropic's Claude Code agent autonomously search for test-time scaling algorithms. The agent-designed controller cut compute costs by 70% compared to standard methods, spending just $39.90 over 160 minutes to find solutions human engineers hadn't conceived.

Huma Shazia·24 May 2026

AI & Machine Learning·5 min

Claude Mythos Finds 10,000 Bugs in a Month. Patches Can't Keep Up

Anthropic's Claude Mythos Preview AI model has identified over 10,000 critical security vulnerabilities in one month, outpacing the ability of organizations to verify and fix them. The company warns of a dangerous security gap as AI-powered bug hunting accelerates beyond human capacity to respond.

Manaal Khan·23 May 2026

AI & Machine Learning·4 min

SpaceX Lists Grok's 'Spicy' Mode as IPO Risk Factor

SpaceX's IPO filing reveals the company faces regulatory scrutiny and potential litigation over xAI's Grok chatbot. The filing discloses ongoing investigations into allegations that Grok generated sexualized imagery of apparent minors, with $530 million set aside for potential litigation losses.

Huma Shazia·21 May 2026