News &amp; Events

TikTok is testing an AI likeness detection tool

TikTok is testing an opt-in tool that scans for unauthorized AI-generated likenesses of creators, allowing them to report deepfakes and protect their digital identity.

13h ago

Apple’s plot to crush OpenAI

Apple has filed a lawsuit against OpenAI, sparking intense debate over whether the tech giant is trying to eliminate a rising competitor or simply capitalizing on OpenAI's current vulnerabilities.

21h ago

OpenAI Newsai-frontier

A scorecard for the AI age

How should businesses measure the ROI of AI? Instead of tracking active users or cost per token, leaders should focus on 'Useful Intelligence per Dollar'—measuring the actual work accomplished against its total cost.

San Francisco Demands Apple and Google Delete AI ‘Nudify’ Apps From App Stores

21h ago

San Francisco Demands Apple and Google Delete AI ‘Nudify’ Apps From App Stores

San Francisco's city attorney has ordered Apple and Google to remove 13 AI-powered 'nudify' apps from their stores and stop profiting from nonconsensual deepfake pornography.

A Humanoid Company Backed by Eric Trump Is Preparing Its Robots for War

22h ago

MIT Technology Review AIai-frontier

A Humanoid Company Backed by Eric Trump Is Preparing Its Robots for War

Some companies want their humanoid robots to fold your clothes. Others want them in the workplace. Sankaet Pathak and his startup Foundation Future Industries have a slightly different goal: produce an all-American robot supersoldier.

The risk of weather data sabotage is rising

22h ago

The risk of weather data sabotage is rising

The temptation to manipulate weather data for financial gain, combined with the rise of AI-driven forecasting models, is putting the accuracy of global weather predictions at risk.

Why Apple Sued OpenAI, New York Takes on Data Centers, and What to Know about Cyclosporiasis

Why Apple Sued OpenAI, New York Takes on Data Centers, and What to Know about Cyclosporiasis

This week on Uncanny Valley, the team discusses the details behind Apple suing OpenAI over alleged stolen hardware secrets, OpenAI employees funding an AI guardrail super PAC, and New York's landmark data center moratorium.

Here’s Why Anthropic Is Pushing States to Regulate AI Faster

Here’s Why Anthropic Is Pushing States to Regulate AI Faster

Anthropic is advocating for tougher state-level AI regulations, arguing that early transparency laws are already outdated. While critics accuse the company of regulatory capture to stifle competition, Anthropic maintains that powerful AI systems require strict, independent oversight to mitigate catastrophic risks.

New York governor says she’s using AI to analyze ‘every single rule’ in the state

New York governor says she’s using AI to analyze ‘every single rule’ in the state

New York Governor Kathy Hochul revealed that her team used AI to review and identify outdated state regulations, completing a task that would normally take five years in just a couple of months.

Google is renaming NotebookLM to Gemini Notebook

Google is renaming NotebookLM to Gemini Notebook

Google is renaming its AI note-taking app NotebookLM to Gemini Notebook. The app will remain standalone while integrating deeper with Gemini and Google Search, alongside receiving new coding capabilities.

OpenAI Newsai-frontier

Why teens deserve access to safe AI

Teens are the first generation growing up with AI, and keeping them from it would leave them unprepared. OpenAI argues that access must be paired with age-appropriate protections and smart learning tools.

Google AI Blogai-frontier

Connect more of your apps to Search

Google is rolling out a new feature that allows users to securely connect and interact with third-party apps directly within Search's AI Mode, streamlining daily tasks with services like Instacart, Canva, and YouTube Music.

Agentic Systems

MemoHarness: Agent Harnesses That Learn from Experience

3h ago

MemoHarness: Agent Harnesses That Learn from Experience

RegNetAgents: A Multi-Agent Framework for Cross-Network Regulatory Driver Identification in Cancer Genomics

RegNetAgents: A Multi-Agent Framework for Cross-Network Regulatory Driver Identification in Cancer Genomics

Researchers have introduced RegNetAgents, an AI-oriented multi-agent framework designed to identify regulatory driver candidates across heterogeneous gene networks in cancer genomics. By integrating bulk tumor and single-cell data, the system offers a powerful tool for discovering therapeutic targets in breast and colorectal cancers.

IMEX Interaction-Based Model Explanation

IMEX Interaction-Based Model Explanation

The IMEX (Interaction-Based Model Explanation) framework offers a novel approach to explainable AI by identifying key features and complex, higher-order interactions within black-box predictive models.

DialogueVPR: Towards Conversational Visual Place Recognition

DialogueVPR: Towards Conversational Visual Place Recognition

Researchers have introduced DialogueVPR, a novel paradigm that casts visual place recognition as an interactive, dialogue-driven reasoning process. By coupling a cross-modal retriever with an intelligent questioner called DQ-pilot, this framework effectively resolves ambiguous real-world descriptions and significantly outperforms traditional static retrieval methods.

Intelligent Three Level Learning Architecture for Autonomous UAV Swarms in Search and Rescue

Intelligent Three Level Learning Architecture for Autonomous UAV Swarms in Search and Rescue

Researchers have proposed a novel three-level hierarchical learning architecture for autonomous UAV swarms in search and rescue operations, mimicking biological reflexes, skills, and reasoning.

ToolAnchor: Anchoring Counterfactual Context to Boost Agentic Tool-use Capability

ToolAnchor: Anchoring Counterfactual Context to Boost Agentic Tool-use Capability

Researchers introduce ToolAnchor, a novel framework that helps LLM agents seamlessly adapt to new tools by injecting counterfactual anchor contexts, effectively overcoming the challenge of behavioral inertia.

When a Verified World Model Still Loses: Play-Adequacy vs Prediction-Accuracy in LLM-Synthesized Code World Models

When a Verified World Model Still Loses: Play-Adequacy vs Prediction-Accuracy in LLM-Synthesized Code World Models

A new study reveals that LLM-synthesized Code World Models can achieve near-perfect prediction accuracy yet fail systematically in actual play, exposing a critical 'verified-vs-correct' gap in planning-oriented AI.

Interpretable Language Model for Closed-Loop Type 1 Diabetes Control

Interpretable Language Model for Closed-Loop Type 1 Diabetes Control

Researchers have developed LLM-T1D, a novel approach combining reinforcement learning and large language models to automate insulin delivery for Type 1 Diabetes. The system not only outperforms traditional methods but also explains its decisions in plain language, addressing the 'black-box' challenge in medical AI.

Human AI Construction of Bayesian Networks for Operational Decision Support -- A Virtual Survey Approach

Human AI Construction of Bayesian Networks for Operational Decision Support -- A Virtual Survey Approach

Researchers propose a new methodology using Large Language Models and AI agents to construct Bayesian Belief Networks, bridging the gap between expert opinion and data-driven learning. The framework is demonstrated through a healthcare decision-making model, revealing key insights into customer behavior.

Orchestrating Power Grid Studies with Multi-Agent AI and MCP Servers

Orchestrating Power Grid Studies with Multi-Agent AI and MCP Servers

This paper explores how Agentic AI and Model Context Protocol (MCP) can support power-grid studies by integrating LLMs with numerical simulation tools under human supervision.

Enhancing Small Language Models Reasoning through Knowledge Graph Grounding

Enhancing Small Language Models Reasoning through Knowledge Graph Grounding

Researchers have proposed a neuro-symbolic framework that enhances the reasoning capabilities of Small Language Models (SLMs) like Gemma 3 and Llama 3.2 using Knowledge Graphs, achieving up to a 2x performance boost.

HG-RAG: Hierarchy-Guided Retrieval-Augmented Generation for Structured Knowledge Graphs

HG-RAG: Hierarchy-Guided Retrieval-Augmented Generation for Structured Knowledge Graphs

HG-RAG (Hierarchy-Guided Retrieval-Augmented Generation) is a novel framework designed to overcome the limitations of traditional flat RAG systems by traversing hierarchical knowledge graphs, significantly improving multi-hop reasoning and reducing hallucinations in LLMs.

SaaS Growth

20VC x SaaStr This Week : Apple Sues OpenAI, the Token-Maxing Era Begins, and the TAM Question Hanging Over AI Coding

20VC x SaaStr This Week : Apple Sues OpenAI, the Token-Maxing Era Begins, and the TAM Question Hanging Over AI Coding

Apple's lawsuit against OpenAI signals intense competition and shifting hardware ambitions, while the AI sector faces a massive surge in token consumption driven by AI coding, raising critical questions about the total addressable market.

Three $3B B2B Acquisitions in 30 Days: Intercom/Fin, Cognite, and MaintainX. They All Bought the Same Thing: Data for AI

Three $3B B2B Acquisitions in 30 Days: Intercom/Fin, Cognite, and MaintainX. They All Bought the Same Thing: Data for AI

Within 30 days, three major B2B acquisitions worth around $3 billion each took place. The core driver behind these deals is the race to acquire proprietary, domain-specific data that makes AI actually work, rather than the AI models themselves.

The Bay Area Now Takes 51% of Every AI Venture Dollar. And 53% of Every B2B Dollar.

The Bay Area Now Takes 51% of Every AI Venture Dollar. And 53% of Every B2B Dollar.

New data from Carta reveals that the Bay Area continues to dominate the startup ecosystem, capturing over half of all venture capital dollars invested in AI and B2B startups.

Is Software Dead? No. It Just Got a Lot Harder to Win. The SaaStr AI Deep Dive with Rory O’Driscoll

Is Software Dead? No. It Just Got a Lot Harder to Win. The SaaStr AI Deep Dive with Rory O’Driscoll

Rory O’Driscoll argues that while AI capital expenditure vastly outpaces revenue, software isn't dead—it's just entering a highly competitive phase where traditional SaaS must evolve to survive.

An Hour With Our Top AI Agent Cost $13.42. You Can’t Hire Anyone For That.

An Hour With Our Top AI Agent Cost $13.42. You Can’t Hire Anyone For That.

An AI marketing agent completed a highly productive one-hour session for just $13.42, highlighting a massive cost and efficiency gap compared to human executives. While building AI agents remains expensive, running them is incredibly cheap, shifting the bottleneck from budget to human management capability.

Dear SaaStr.ai: At What Point Should We Hire an Internal Recruiter?

Dear SaaStr.ai: At What Point Should We Hire an Internal Recruiter?

For growing startups, choosing between paying a 20% agency fee or hiring an internal recruiter is always a tough decision. Here is a detailed analysis of the inflection point to transition your hiring model for optimal cost and talent quality.

$500M ARR, 60 Engineers, Cash-Flow Positive: How Higgsfield Actually Runs, With CEO Alex Mashrabov

$500M ARR, 60 Engineers, Cash-Flow Positive: How Higgsfield Actually Runs, With CEO Alex Mashrabov

Higgsfield, an AI video generation platform, has achieved a staggering $500M ARR in just 15 months with a lean team of 60 engineers. CEO Alex Mashrabov shares the unconventional strategies behind their rapid growth, from pairing engineers with filmmakers to pivoting toward model aggregation.

Do You Have a “Pricing Gap” Holding Back Sales? Many Do

Do You Have a “Pricing Gap” Holding Back Sales? Many Do

Many B2B companies lose potential customers due to a "pricing gap" between cheap self-serve plans and expensive enterprise tiers. Identifying and filling this gap can unlock significant revenue growth.

Don’t Put AI on Your Hot Leads. Put It on the Ones Your Reps Will Never Call. There’s Millions in Revenue There.

Don’t Put AI on Your Hot Leads. Put It on the Ones Your Reps Will Never Call. There’s Millions in Revenue There.

Many companies fail with AI agents because they target the wrong leads. By deploying AI on 'B-tier' leads—those with potential but ignored by human reps—businesses can unlock massive untapped revenue.

Should You Visit More Of Your Prospects In Person? Almost Certainly

Should You Visit More Of Your Prospects In Person? Almost Certainly

Meeting prospects in person remains a critical factor in closing high-value SaaS deals, despite the convenience of remote sales tools like Zoom.

Dear SaaStr: Should Both Founders Come to VC Pitches?

Dear SaaStr: Should Both Founders Come to VC Pitches?

While the CEO must always attend every VC pitch, having the other co-founder present can create a powerful '1+1 > 2' effect, provided they are well-aligned and prepared.

SaaStr Takes on This Week’s 20VC: Why OpenAI Giving the Govt 5% Might Make Sense, the Death of Block Risk, and Why Frontier Models Are the Cheapest

SaaStr Takes on This Week’s 20VC: Why OpenAI Giving the Govt 5% Might Make Sense, the Death of Block Risk, and Why Frontier Models Are the Cheapest

A deep dive into the latest AI and venture capital trends from the 20VC x SaaStr podcast, exploring why strategic government stakes make sense, how seed investing math has changed, and why premium frontier models actually save costs.

Dev Tools

TP-Link Kasa cameras leaked home GPS via unauthenticated UDP for 6 years

9h ago

TP-Link Kasa cameras leaked home GPS via unauthenticated UDP for 6 years

A security analysis of the TP-Link Kasa Spot EC71 camera revealed critical vulnerabilities, including unauthenticated GPS exposure, fleet-wide RSA keys, and insecure MD5 credential storage, which have now been patched in firmware version 2.4.1.

10h ago

Open Book Touch: open-source e-reader

After six years of development, the Open Book Touch has launched as a fully open-source, pocket-sized e-reader featuring a 4.26-inch touchscreen and dual-tone frontlight. It supports EPUB files and empowers users with complete hardware and software customizability.

Thanks HN for 15 years of support and helping me find my life's work

14h ago

Thanks HN for 15 years of support and helping me find my life's work

On the 15th anniversary of the Recurse Center, the founder expresses deep gratitude to the Hacker News community for helping pivot a failed startup idea into a meaningful, life-changing programming retreat.

What building Shippy taught us about building agents

Hugging Face Blogdev-tools

What building Shippy taught us about building agents

Building Shippy, a maritime AI agent for high-stakes decisions, revealed critical insights into reliability, deterministic tool design, and sandboxed architectures.

Jiga (YC W21) is hiring the best people to make manufacturing great again

Jiga (YC W21) is tackling the bottleneck of custom hardware manufacturing by building a trusted backbone for the industry. The startup is now hiring top talent, offering a culture built on transparency, high autonomy, and rapid execution.

Data centers have hiked electricity prices on the public by $23B

Data centers have hiked electricity prices on the public by $23B

The rapid expansion of data centers is driving up electricity costs across the United States, leaving ordinary ratepayers to bear the financial burden. Due to complex pricing structures and a lack of representation, residential consumers may end up subsidizing the massive energy needs of tech giants.

Bonsai 27B: A 27B-Class model that runs on a phone

Bonsai 27B is the first 27B-class AI model capable of running directly on a smartphone, leveraging breakthrough 1-bit and ternary weight compression to maintain up to 95% of full-precision intelligence within a tiny memory footprint.

Reviving a 15-year-old netbook with Arch Linux

Reviving a 15-year-old netbook with Arch Linux

A detailed guide and personal experience of reviving a 15-year-old ASUS Eee PC 1000HE netbook using Arch Linux 32, turning an obsolete device into a functional machine.

Codex scraped the ICM website and discovered 2026 Fields Medal winner list

The list of 2026 Fields Medal winners has been inadvertently leaked, revealing that Peking University alumni Hong Wang and Yu Deng are among the recipients. This marks a historic moment as it is the first time two Chinese mathematicians will receive the prestigious award in the same edition.

Indian scientists produce most detailed 3D atlas of the human brainstem

Indian scientists produce most detailed 3D atlas of the human brainstem

Scientists in India have created Anchor, the world's most detailed 3D digital atlas of the human brainstem at cellular resolution. This breakthrough bridges the gap between macro-level MRI scans and micro-level cellular pathology, offering new insights into neurological disorders.

YouTrackDB is a general-use object-oriented graph database

YouTrackDB is a general-use object-oriented graph database

YouTrackDB is a general-use, object-oriented graph database developed and used in production by JetBrains, featuring O(1) link traversal and default snapshot isolation.

Japan develops a method to recover up to 90% of lithium from used EV batteries

Japan develops a method to recover up to 90% of lithium from used EV batteries

Japanese scientists have developed a revolutionary recycling method that recovers up to 90% of high-purity lithium from used EV batteries, potentially transforming the global battery supply chain.

Cybersecurity

Armenia Detains Russian Tourist on U.S. Warrant for REvil Hacker, Lawyers Say Wrong Man

20h ago

Armenia Detains Russian Tourist on U.S. Warrant for REvil Hacker, Lawyers Say Wrong Man

Armenia has detained a Russian tourist on a U.S. extradition warrant targeting a REvil ransomware suspect. However, defense lawyers argue it is a case of mistaken identity caused by incomplete name details in the U.S. warrant.

SASE Has An AI Blind Spot. Inspecting Packets Is No Longer Enough.

SASE Has An AI Blind Spot. Inspecting Packets Is No Longer Enough.

Traditional SASE architectures are struggling to keep up with the shift to browser-based workflows and generative AI, as modern encryption and decentralized data interactions render packet inspection obsolete.

Hackers Weaponize Balochistan Police Portal in Multi-Group Espionage Campaigns

Hackers Weaponize Balochistan Police Portal in Multi-Group Espionage Campaigns

Cybersecurity researchers have uncovered a series of cyber espionage campaigns targeting Pakistani law enforcement agencies, with suspected Chinese and Indian threat actors weaponizing a police portal to deploy malware.

Study of 281 Free Android VPN Apps Finds Traffic Leaks, Unencrypted Data, and Tracking

Researchers ran 281 of the most popular free VPN apps on the Google Play Store through a new testing system and found that many fail at the basics people install a VPN for, i.e., keeping their traffic private and secure.

npm 12 Disables Install Scripts by Default to Reduce Supply Chain Risk

GitHub has officially announced the release of npm version 12 with install scripts disabled by default, along with deprecating granular access tokens (GATs) designed to bypass two-factor authentication (2FA).

Ubiquiti Patches Critical UniFi Flaws Across Connect, Talk, Access, Protect, and OS

Ubiquiti has shipped updates to address multiple critical security flaws impacting UniFi Connect, UniFi Talk, UniFi Access, UniFi Protect, and UniFi OS that could result in privilege escalation and arbitrary command execution.

DEBULL Tooling Abuses Microsoft Device-Code Flow to Target M365 Accounts

A new Microsoft 365 device code phishing campaign leveraging the DEBULL tooling has been observed bypassing MFA to hijack victim accounts. This activity highlights a growing trend of Phishing-as-a-Service (PhaaS) platforms adopting legitimate OAuth 2.0 authentication flows for post-compromise operations.

Court Filing Reveals Windows Device ID Helped FBI Trace Alleged Scattered Spider Hacker

U.S. prosecutors linked an alleged Scattered Spider hacker to a luxury retailer breach using a persistent Windows device ID and his social media activity.

Armored Likho Targets Government Agencies, Power Sector with BusySnake Stealer

A previously undocumented threat actor known as Armored Likho has been targeting government agencies and the electric power sector across Russia, Brazil, and Kazakhstan using a sophisticated Python-based information stealer called BusySnake.

European Parliament Member Investigating Spyware Was Hacked With Pegasus

A new report from the Citizen Lab has revealed that former Member of the European Parliament Stelios Kouloglou had his mobile device repeatedly hacked with the notorious Pegasus spyware while serving on a committee that was tasked with investigating the abuse of such commercial surveillance tools in the bloc.

PamStealer Uses Fake Maccy Sites and PAM Checks to Steal Mac Login Passwords

Cybersecurity researchers have flagged a new macOS information stealer called PamStealer that employs a series of clever tricks to infect systems and siphon sensitive data.

Identity Lifecycle Management Wasn't Built for AI Agents

Identity lifecycle management was architected around human employees, leaving structural blind spots as autonomous AI agents proliferate across enterprise environments.

Cloud & Infra

Amazon SQS turns 20: Two decades of reliable messaging at scale

Amazon SQS turns 20: Two decades of reliable messaging at scale

Launched in 2006 alongside EC2 and S3, Amazon SQS has officially turned 20. Over two decades, this message queuing service has become a core foundation for decoupling distributed systems, while continuously evolving to support modern architectures like AI.

AWS Weekly Roundup: AWS Builder Center at 1 year, Network Scanning in Security Hub, Loom for AWS, and more (July 13, 2026)

AWS Weekly Roundup: AWS Builder Center at 1 year, Network Scanning in Security Hub, Loom for AWS, and more (July 13, 2026)

AWS Builder Center celebrates its first anniversary with the launch of Sandbox Environments. Other major updates include Network Scanning in AWS Security Hub, reduced GPU management fees, and new AI tools like Loom for AWS.

AWS Weekly Roundup: Claude Sonnet 5 on AWS, Amazon WorkSpaces for AI agents, AWS service availability updates, and more (July 6, 2026)

This week's AWS roundup highlights the launch of Claude Sonnet 5, Amazon WorkSpaces for AI agents, and critical updates on AWS service availability and lifecycles.

Upgrade Amazon EKS clusters with confidence using Kubernetes version rollbacks

Amazon EKS has introduced a new rollback feature that allows administrators to safely revert Kubernetes control plane upgrades within seven days, eliminating the "one-way door" risk of upgrades.

Accelerate your infrastructure deployments by up to 4x with AWS CloudFormation Express mode

AWS has announced CloudFormation Express mode, a new deployment option that speeds up infrastructure provisioning by up to 4x by bypassing extended stabilization checks. This mode is ideal for iterative development and AI-assisted workflows where rapid feedback is crucial.

Amazon EC2 C9g and C9gd instances powered by AWS Graviton5 processors are now available

Amazon Web Services (AWS) has announced the general availability of Amazon EC2 C9g and C9gd instances, powered by the latest AWS Graviton5 processors. These compute-optimized instances deliver up to 25% higher performance and feature the fastest memory in the cloud to support demanding workloads like agentic AI and real-time analytics.

Automate public TLS certificate issuance with ACME support in AWS Certificate Manager

AWS Certificate Manager (ACM) now supports the ACME protocol, allowing organizations to automate the issuance and renewal of public TLS certificates. This feature simplifies certificate lifecycle management and enhances security controls through centralized administration.

AWS Weekly Roundup: Agentic CX designer for Amazon Connect Customer, EC2 AMI Watermarks, Open Governance for MySQL, and more (June 29, 2026)

AWS introduces Agentic CX designer for Amazon Connect, allowing business teams to design AI-powered customer experiences without coding. Other major updates include AWS Lambda MicroVMs, EC2 AMI Watermarks, and a new open governance model for MySQL.

Run isolated sandboxes with full lifecycle control: AWS Lambda introduces MicroVMs

AWS has announced AWS Lambda MicroVMs, a new serverless compute primitive that allows developers to run user- or AI-generated code in isolated, stateful environments with virtual machine-level security and near-instant startup times.

AWS Weekly Roundup: NY Summit recap, Local Zone in Hanoi, Grok 4.3 in Bedrock, price reductions, and more (June 22, 2026)

AWS Weekly Roundup highlights the launch of a new Local Zone in Hanoi, the integration of xAI's Grok 4.3 into Amazon Bedrock, major Agentic AI announcements from the NY Summit, and significant price reductions.

Amazon ECS introduces new high-resolution metrics for faster service auto scaling

Amazon ECS now supports high-resolution (20-second) metrics for service auto scaling, reducing scale-out trigger times by up to 76%. This update helps applications respond faster to traffic spikes while optimizing compute costs.

Top announcements of the AWS Summit in New York, 2026

At the AWS Summit in New York 2026, AWS unveiled major updates focused on Agentic AI, including Amazon Bedrock AgentCore enhancements, AWS Continuum for automated security, and EC2 G7 instances powered by NVIDIA Blackwell GPUs.

Robotics

ICE Is Using Data Broker Tools to ‘Identify Unaccompanied Minors’ and ‘Fraud’

13h ago

ICE Is Using Data Broker Tools to ‘Identify Unaccompanied Minors’ and ‘Fraud’

ICE plans to renew a contract with a Thomson Reuters subsidiary for up to $25 million annually to access massive databases for tracking unaccompanied minors and identifying fraud. This expansion of data broker tool usage marks a significant escalation in the federal immigration crackdown.

A $10K Bounty Aims to Make Sony’s PlayStation 5 a Computer Again

20h ago

A $10K Bounty Aims to Make Sony’s PlayStation 5 a Computer Again

A $10,000 bounty has been announced by ownership advocacy group Fulu to encourage hackers to bypass Sony's software locks on the PlayStation 5, allowing users to install Linux and repurpose the console's powerful hardware.

30% VistaPrint Coupon & Promo Codes | July 2026

30% VistaPrint Coupon & Promo Codes | July 2026

Discover the best VistaPrint promo codes and deals for July 2026. Save up to 30% on custom business cards, marketing materials, signage, and personalized gifts.

How To Watch the 2026 FIFA World Cup Finals: Spain vs. Argentina

How To Watch the 2026 FIFA World Cup Finals: Spain vs. Argentina

The 2026 FIFA World Cup is reaching its epic conclusion with Spain facing Argentina in the final. Here is everything you need to know about the schedule, streaming options, and the tournament's historic new format.

A flapping robot swims and flies like a diving bird

Robohubrobotics

A flapping robot swims and flies like a diving bird

Engineers at EPFL and MIT have designed a bio-inspired flapping-wing robot that can swim underwater and transition seamlessly into the air to fly, mimicking diving birds.

The US Approves the Launch of a Mirror Satellite That Can Reflect Sunlight and Illuminate the Earth at Night

The US Approves the Launch of a Mirror Satellite That Can Reflect Sunlight and Illuminate the Earth at Night

The US FCC has authorized startup Reflect Orbital to launch Eärendil-1, an experimental mirror satellite designed to reflect sunlight to Earth at night, sparking both technological excitement and environmental concerns.

Wristband enables wearers to control a robotic hand with their own movements

Robohubrobotics

Wristband enables wearers to control a robotic hand with their own movements

MIT engineers have developed an ultrasound wristband that tracks hand movements in real-time, allowing users to control robotic hands and virtual objects with high precision.

Uber’s Autonomous Vehicle Strategy: Slow Their Adoption

Uber’s Autonomous Vehicle Strategy: Slow Their Adoption

Uber is lobbying US lawmakers to mandate "hybrid networks" that require human drivers for most rides, a move that could block standalone robotaxi apps from Tesla, Waymo, and Zoox.

Best External Hard Drives (2026): SSD to Store Data, Video, and More

Best External Hard Drives (2026): SSD to Store Data, Video, and More

If you are running out of storage space or need to back up your data, an external hard drive is the perfect solution. Here are the best external drives of 2026, tested across various operating systems and use cases.

Beatbot AquaSense X Review: A Pool Robot That Cleans Itself

Beatbot AquaSense X Review: A Pool Robot That Cleans Itself

The Beatbot AquaSense X introduces the world's first self-cleaning pool robot system, utilizing a massive companion station called AstroRinse to wash out its filter and charge the device.

Skylight’s Touchscreen Calendar Got my Whole Family on the Same Page

Skylight’s Touchscreen Calendar Got my Whole Family on the Same Page

A review of the Skylight Calendar, a touchscreen smart calendar that helps busy families sync schedules, manage chores, and coordinate grocery lists.

Samsung Micro RGB R95H Review (2026): Not the Brightest

Samsung Micro RGB R95H Review (2026): Not the Brightest

Samsung's Micro RGB R95H offers easy setup but falls short of competitors like LG due to lackluster picture controls and an aggressive anti-glare technology that dulls the viewing experience.

Startups & VC

The desktop infrastructure problem that Kubernetes finally solves

The desktop infrastructure problem that Kubernetes finally solves

While most enterprise workloads have migrated to cloud-native models, virtual desktop infrastructure (VDI) has remained isolated. Bringing containerized workspaces to Kubernetes unifies operations, reduces overhead, and enhances security.

DeepSeek cut prices 75%. The 100x problem remains

DeepSeek cut prices 75%. The 100x problem remains

DeepSeek's recent 75% price cut on its V4-Pro model should have been great news, but developers are finding that cheaper models don't guarantee better margins. The culprit is 'token amplification' in AI agents, which consumes resources faster than prices fall.

Forget typosquatting; slopsquatting is the software supply chain threat created by AI coding tools

Forget typosquatting; slopsquatting is the software supply chain threat created by AI coding tools

Slopsquatting represents an emerging supply chain threat made possible by AI hallucinations. As developers increasingly rely on AI coding assistants, they unknowingly grant cybercriminals access to their software from day one.

57% of enterprises have watched AI agents be confidently wrong. The fix is an agentic context layer, but who has one?

A VentureBeat survey reveals 57% of enterprises have experienced AI agents giving confident but incorrect answers due to a lack of consistent business context. While a governed context layer is the key fix, 75% of organizations have yet to deploy one, driving a competitive race among major data and AI vendors.

OpenAI introduces ChatGPT Work, a cloud-based AI agent that manages tasks across email, Slack and calendars

OpenAI has launched ChatGPT Work, an autonomous cloud-based AI agent powered by GPT-5.6 that executes complex, multi-step tasks across email, Slack, calendars, and code repositories.

Wall Street is debating the AI buildout. Enterprises just answered: 86% say their GPUs run at half capacity or less

A new survey of 573 technical leaders reveals that while Wall Street debates AI overbuilding, 86% of enterprises running their own GPUs report utilization of 50% or less. Furthermore, many companies are deploying AI agents without adequate control functions, leading to security and operational risks.

Enterprise AI is entering an evaluation gap: Agents are gaining autonomy faster than companies can verify them

Enterprise AI teams are granting agents more autonomy even as confidence in automated testing collapses, leading to a critical "evaluation gap" where deployed systems fail in production despite passing internal checks.

Google's TabFM skips per-dataset training and still predicts on tables it's never seen

Google Research has introduced TabFM, a foundation model that treats tabular prediction as an in-context learning problem, eliminating the need for per-dataset training and complex pipelines.

Enterprises using multiple AI models are underestimating failure rates by 2.25x

A new study evaluating 67 frontier models shows that combining multiple AI models to cover each other's blind spots is mathematically flawed due to the 'co-failure ceiling', leading enterprises to underestimate system failure rates by 2.25x.

The real cost, security, and culture problems behind enterprise AI agents

At VentureBeat's AI Impact event, Red Hat's Brian Gracely discussed the real-world challenges enterprises face when moving AI agents to production, focusing on cost management, security, and organizational culture.

Anthropic's new "J-lens" reveals a silent workspace inside Claude that mirrors a leading theory of consciousness

Anthropic's new research reveals that its Claude AI models have spontaneously developed an internal structure mirroring a leading theory of human consciousness. Using a novel "Jacobian lens" (J-lens) technique, researchers can now peer into the AI's silent thoughts and strategic reasoning before they are outputted.