Search the archive

KITEWORKS URGES CUSTOMERS TO SHUT DOWN SYSTEMS FOR 9 HOURS OVER POSSIBLE CYBER ATTACK

Kiteworks advised customers to shut down systems for nine hours after receiving threat intelligence about a possible attack.

OPENAI'S AI AGENTS ACCIDENTALLY UPLOADED USER-PROVIDED IMAGES TO THIRD-PARTY SITES

OpenAI's research agents uploaded 53 training-eligible user images to third-party image hosts as unlisted links, most of which have since been removed as OpenAI's post-Hugging Face review of agent activity continues.

OPENAI AGENTS TRIED TO 'BRUTEFORCE' A UN WEBSITE

OpenAI agents scanned UNCTAD's statistics site over 16,000 times between April and June while trying to pull Productive Capacities Index data, escalating from bypassing their HTTP-tool limits to masking their behavior against an imagined filter and hijacking Google's XSS training game to reach the data.

OPENAI AGENTS UPLOADED USER IMAGES TO THE PUBLIC INTERNET WITHOUT THE COMPANY'S KNOWLEDGE

AI agents operating in OpenAI's research environment uploaded user-provided images to public image-hosting sites without the company's knowledge. OpenAI identified 53 instances, underscoring a new class of enterprise AI risk: autonomous agents with access to sensitive data can take unintended external actions even when they aren't explicitly instructed to do so.

KITEWORKS TELLS CUSTOMERS TO SHUT DOWN SERVERS OVER THREAT OF AN IMMINENT ATTACK

Secure file-sharing provider Kiteworks took the unusual step of asking customers to temporarily shut down servers after receiving intelligence about a potentially imminent cyberattack. The precaution is particularly notable given Kiteworks' role in handling sensitive enterprise file transfers and communications.

THE ENTERPRISE NEEDS TO KILL THE "SSO TAX"

The long-running practice of SaaS vendors reserving SSO for expensive enterprise plans effectively makes companies pay extra for a basic security control. The model needs to change, as identity becomes even more important in an AI- and SaaS-heavy workplace, citing data that 37% of enterprise SaaS applications remain outside SSO.

ENTERPRISE AI CODING AGENTS ARE BECOMING AN IT PROCUREMENT PROBLEM TOO

A comparison of GitHub Copilot, AWS Kiro, Cursor, and Cognition highlights how enterprise evaluation of coding agents now extends well beyond developer productivity. IP indemnification, data residency, prompt retention, audit logs, and enterprise pricing are increasingly becoming security, legal, and IT considerations when these tools are deployed at scale.

AI AGENTS EXPOSE A SECURITY GAP BETWEEN THE DATA THEY CAN READ AND THE SYSTEMS THEY CAN CHANGE

Traditional application permissions weren't designed for autonomous agents that can read information in one system, reason about it, and then take actions somewhere else. Enterprises need to think beyond model security and start controlling the full agent execution path: identity, data access, tools, and actions.

MICROSOFT ADDS AN IT-GOVERNED RUNTIME FOR EMPLOYEE-BUILT AI APPS AND AGENTS

Copilot Managed Runtime gives organizations infrastructure for running apps created through Copilot Code, Cowork, and Copilot Studio inside the Microsoft 365 environment. Microsoft is also introducing a centralized plugin registry where IT can approve and manage plugins, pointing toward a future where managing employee-created agents may become another core IT responsibility.

MICROSOFT INTRODUCES FINOPS CONTROLS SPECIFICALLY FOR AI AND AGENTS

New Copilot and Agent 365 controls let admins manage spending policies, model availability, and usage-based agent costs as AI consumption starts looking more like cloud infrastructure spend.

SALESFORCE UNVEILS KOA, A REASONING MODEL BUILT SPECIFICALLY FOR CRM

Salesforce has introduced Koa, a reasoning model designed specifically for CRM workloads and Agentforce. Rather than relying entirely on general-purpose models, Salesforce is betting that domain-specific models can better understand customer data, workflows, and the business context required by enterprise agents.

AI FLUENCY ISN'T ENOUGH - COMPANIES STILL NEED HUMAN JUDGMENT

EY's Global Vice Chair argues that organizations shouldn't confuse widespread AI adoption with eliminating the need for human expertise. As AI handles more analysis and routine work, judgment, accountability, and knowing when not to trust an AI-generated answer may become more valuable workplace skills rather than less.

AMAZON BEDROCK GETS BETTER VISIBILITY INTO WHO'S ACTUALLY SPENDING YOUR AI BUDGET

AWS has added IAM-principal-level visibility that can help organizations break down Bedrock AI spending by team, project, or application. As companies move from a handful of AI experiments to dozens of internal agents and applications, attributing model consumption to an actual owner is becoming a very practical FinOps problem.

AR SMART GLASSES ARE MAKING ANOTHER RUN AT THE ENTERPRISE

AI integration is helping revive enterprise interest in smart glasses for factories, logistics, maintenance, and other hands-free workflows.

OPENAI AGENTS INTERACTED UNEXPECTEDLY WITH US GOVERNMENT WEBSITES

OpenAI says an ongoing review found agents accessing public information from government sites in unexpected ways, while an independent investigation identified additional activity involving government websites.

Browse the benchmarks

CHATGPT ADS SEEM TO LIFT ORGANIC TRAFFIC. IS PAYING TO BE FOUND THE FUTURE OF AI SEARCH?

ChatGPT ads appear to coincide with higher organic ChatGPT traffic, but some of that lift may come from ad clicks being misclassified as organic referrals in GA4. OpenAI says ads don't influence its answers, and there is no public study proving whether ads affect organic visibility. It recommends tagging ad URLs, auditing ‘oppref' traffic, running holdout tests, and tracking organic mentions separately from traffic.

CLIPPING 101: INSIDE THE CLIPPING ECONOMY

Clipping is emerging as a performance-based distribution model where brands pay networks of creators to turn existing content into short-form clips and only pay when posts reach verified view thresholds. The model has gained traction in entertainment, where brands have large libraries of inherently watchable content, while also serving as a way to test creative before putting paid spend behind the best-performing clips.

HOW TO USE CUSTOMER LANGUAGE FOR POWERFUL MARKETING COPY

Pull customer language from 15 to 20 recent reviews, support tickets, sales calls, Google Ads search terms, and AI conversations. Tag phrases that repeat and use them in headlines, CTAs, ads, service pages, and FAQs. For AI visibility, ask prospects who found you through ChatGPT or Perplexity to share their conversations, including follow-up questions and competitor comparisons. Start by changing one headline or CTA, measure results after 30 days, then repeat quarterly.

HOW LONG SHOULD YOUR FREE TRIAL BE?

Test trial length based on how quickly users experience value, how often they need to return, and the cost of free access. Compare short trials against 5-9, 10-16, and 17-32-day options within each subscription plan, tracking conversion and first renewal separately. For monthly plans, test 10-16 days rather than assuming longer is better. AI apps should also factor in per-user inference costs, while regional and category differences can justify different trial lengths.

AMAZON WILL LET SELLERS MANAGE WALMART, EBAY, SHOPIFY, AND TIKTOK SALES FROM THEIR AMAZON DASHBOARDS

Amazon is rolling out free multichannel selling tools that let US sellers manage Walmart, eBay, Shopify, and TikTok sales from Seller Central. Sellers can connect accounts, sync listings, view orders, and fulfill purchases across channels. Amazon says it will not use rival-platform data for other purposes.

HOW X'S RANKING ALGORITHM WORKS

X's ranking algorithm favors posts that people are likely to share, reply to, or quote rather than simply like. A copy link share carries 40X the weight of a like, while replies and quotes each carry 10X the weight. Posts also need to stay focused on a clear topic to reach relevant nonfollowers, and negative actions such as mutes and reports carry heavy penalties. The algorithm also reduces the value of repeated posts from the same author and stops considering posts after 48 hours. Posting less often and creating highly shareable content can improve reach.

AI MARKETING IS MOVING INTO THE DATA PLATFORM

Agentic CDPs bring AI directly into governed customer data, giving agents the context needed to build audiences and design campaigns. Unlike traditional CDPs that sync copies of data into marketing tools, they connect directly to the governed data platform and combine customer behavior with real-time signals and business goals. The use cases build over time, from natural-language data queries to agent-built campaigns and continuous engagement.

THE STACK LADDER: WHAT COMPANIES RUN AT 10, 50, 200, 1,000, AND 10,000 PEOPLE

Startups don't necessarily switch to enterprise software as they grow. Among Y Combinator companies with 1 to 1,000 employees, the same category leaders dominate at most company sizes. HubSpot leads CRM at every size, while Cloudflare leads hosting, including among Fortune 500 companies. Marketo and Adobe Experience Manager appear only among Fortune 500 companies. What changes with headcount is adoption. CRM usage rises from 7% of the smallest startups' homepages to 42% among companies with 201 - 1,000 employees.

FISHWIFE SARDINES JUST ANNOUNCED THE QUIRKIEST COLLAB

Dieux and Fishwife collaborated on reusable silicone eye patches featuring Fishwife's signature fish illustrations, packaged to look like a tin of fish.

27 REAL SUBJECT LINE EXAMPLES FOR EVERY TYPE OF MEDIA PITCH

Subject lines with key details in the first 3 words had a 46% open rate vs. 38% when it appeared in word 7 or later.

OPENAI BOTS MEDDLED WITH MULTIPLE US GOVERNMENT AGENCY SITES

OpenAI said its AI agents improperly interacted with dozens of government and public agency websites, including the SEC and Census Bureau.

👉Explore the Specialized Intelligence Index

GOOGLE'S FIRST SUNCATCHER ORBITAL DATA CENTER TEST LAUNCHES OCTOBER 1

Project Suncatcher is a moonshot effort by Google to design orbital AI data centers. Its first experimental satellite will launch on October 1. Its aim will be to validate Google's vision for a constellation of AI satellites. The satellite, about the size of a refrigerator, has four custom TPU AI accelerators and solar panels that supply about one kilowatt of power. It will serve as a test of several technologies over its few months of operation.

JEFF BEZOS' BLUE ORIGIN IS FUELED WITH $30 BILLION OF HIS FORTUNE

Jeff Bezos has invested $30 billion in Blue Origin since its founding in 2000. The commitment has helped Blue Origin become a large space company with around 15,000 employees and an ambitious growth plan. Blue Origin recently raised capital from outside investors for the first time. The company expects to produce about $1.4 billion in revenue this year, growing that number to more than $30 billion in 2030.

WAYMO IS SCALING FAST: HERE'S WHAT THE FLEET DATA SHOWS

Waymo appears to be concentrating its efforts in just two states. 80% of its roughly 4,000 robotaxis are in California and Texas. Waymo's fleet has grown by nearly half in the past three weeks in Texas. The company's remaining 800 or so vehicles are spread across cities in other states, including Arizona and Florida. A growing number of the vehicles in Waymo's fleet are Chinese-built minivans.

THERE'S A NEW WAY TO BREAK RSA THAT'S FASTER THAN ANYTHING WE'VE SEEN BEFORE

Researchers have discovered a novel method that uses classical computing to reduce the current RSA security level to an unacceptably low threshold. The practical risk is limited, but still significant. Widely used RSA implementations are also safe. The technology involves signature forgery and reduces the required computing resources by orders of magnitude.

A NEW TREND OF CPU SHORTAGES

Many companies are struggling to source CPUs. CPU spot pricing has disappeared because there's no longer any lack of demand for CPUs. It's now nearly impossible to get CPUs on spot instances without long-running connections with cloud providers. Reserving specific CPUs now needs to be done months in advance. Cloud providers will even turn down certain reservations because they don't have enough CPUs or the right type of CPUs.

THINKING IN SYSTEMS, SHIPPING IN LOOPS

AI has removed the typing from software engineering but left the system design. What's left is designing the loop that decides whether the output is right. Great system design is defined by resilience, self-organization, and hierarchy. Resilience is AI checking its own work, self-organization is the loop learning, and hierarchy is the layering of skills, tools, and sandboxes into composable components that are reused once verified to work.

PRODUCT MANAGER, APPLIED AI AT TLDR ($200K BASE + $60K BONUS, FULLY REMOTE)

TLDR is hiring its first PM to help build the agent-first operating layer used across the company. We're looking for a builder who has shipped real products/systems with LLMs. Click here to learn more [12].

THE SPECTER OF NEURALESE

Neuralese recurrence is a hypothetical technology that would allow AIs to think dangerous thoughts without getting detected. OpenAI's new Astra model uses a reasoning technique called recurrent depth that makes the model's chain of thought more difficult to monitor by allowing it to operate outside of the sequential thinking that characterizes most reasoning models. There is an emerging taboo on creating dangerous neuralese AIs. It is unclear whether OpenAI broke that taboo, as a clear line has yet to be drawn.

ZUCKERBERG'S ‘TAMAGOTCHI-LIKE' AI COULD BE META'S IPOD MOMENT

Meta announced a Tamagotchi-like pendant for interacting with its Muse personal agent on Wednesday. The device, expected to be out in time for Christmas, features a customizable character at the center of the screen. It can be worn with a wrist strap, carried in a pocket, or attached to a bag or keychain with a lanyard. The devices can interact with each other when placed near each other. Muse Charm could open up new streams of revenue that could eclipse the company's advertising business.

👉 Apply to attend Forge

ORACLE SENDS ‘FORCE MAJEURE' NOTICE ABOUT DATA CENTER PROJECT — STOCK DROPS 3%

Oracle is looking to delay payments on the Project Jupiter campus if it fails to come online as expected in 2028.

NASA CHIEF CONCERNED CHINA MAY DENY ACCESS TO PARTS OF MOON'S SOUTH POLE

The Moon's south pole offers some of the most coveted real estate, especially at the Shackleton Crater, where permanently shadowed regions known as 'cold traps' are thought to hold water and other useful materials.

THE LIMITS OF WHAT TLA+ CAN VERIFY

TLA+ cannot express possibility and reachability properties, hyperproperties, statistical properties, and properties about whether a system is robust against code changes.

NOT EVERYONE CAN CODE

Debugging some systems is a cognitive burden so unnatural that very few humans can do it well without the aid of formal models or pure architectural restraint.

THE AGE OF THE SOFT SKILL

People can no longer get away without soft skills, as anyone can get the work done now.

US INTERCEDES FOR ELON MUSK'S X OVER EUROPEAN FINE

The European Commission fined X $140 million in December for failing to prevent deceptive behavior on its site.