AI News Archive: July 3, 2026 — Part 4
Sourced from 500+ daily AI sources, scored by relevance.
- The shift from black box to glass box in AI translation
AI is transforming global business faster than most organisations can govern it. Few functions illustrate that challenge more clearly than translation. Content that once took weeks now moves in hours. Costs have fallen and multilingual communication now scales at unprecedented speed. But as AI accelerates, visibility is disappearing. Across global organisations, AI increasingly generates multilingual [...]
Score: 58🌐 MovesJul 3, 2026https://www.cityam.com/the-shift-from-black-box-to-glass-box-in-ai-translation/ - Radisson Hotel Group launches hospitality industry-leading AI-powered real-time price matching technology
Radisson Hotel Group launches hospitality industry-leading AI-powered real-time price matching technology The Straits Times
- 10 cases that show Indian courts have an AI hallucination problem
The Supreme Court's July 2026 ruling against AI-generated fake precedents follows a string of cases where fabricated judgments and hallucinated citations reached Indian courts. Here's the complete timeline. The post 10 cases that show Indian courts have an AI hallucination problem appeared first on MEDIANAMA .
Score: 58🌐 MovesJul 3, 2026https://www.medianama.com/2026/07/223-10-cases-ai-hallucination-cases-in-indian-courts/ - The Tech Download: Amazon’s devices chief Panos Panay on tech giant's AI gadget push
CNBC's Arjun Kharpal sits down Amazon's Panay on the latest episode of The Tech Download podcast.
Score: 58🌐 MovesJul 3, 2026https://www.cnbc.com/2026/07/03/the-tech-download-amazon-devices-chief-panos-panay.html - AI is making software developers faster — just not at actually shipping software
A study of more than 100,000 developers finds a vast gap between writing code and shipping software. The reason is human bottlenecks
Score: 57🌐 MovesJul 3, 2026https://qz.com/ai-coding-tools-code-volume-releases-gap-nber-study-061126 - Singapore, AI, and the rise of emotional outsourcing
People used to ask AI for help with parts of modern life, like making an email sound less annoyed or explaining a spreadsheet formula without forcing anyone to revisit their relationship with mathematics. Lately, the exchange has become more intimate. The same tools built to summarise, draft and optimise are now being invited into moments […] The post Singapore, AI, and the rise of emotional outsourcing appeared first on e27 .
Score: 57🌐 MovesJul 3, 2026https://e27.co/singapore-ai-and-the-rise-of-emotional-outsourcing-20260701/ - As AI reshapes ecommerce, Hostinger launches a new platform for small businesses
This AI-assisted dashboard integrates inventory, payments, shipping, and customer data, allowing merchants to manage sales holistically. With competitive pricing and no transaction fees, Hostinger is aggressively targeting India's burgeoning online market, reflecting a broader industry shift towards comprehensive commerce operating systems.
- Cisco’s in-house AI assistant is a jack of all trades
Since the advent of ChatGPT, enterprises have been intent on transforming generative AI’s potential as a digital assistant into productivity enhancements in every pocket of the organization. Networking giant Cisco is one company that has been at the leading edge of that pursuit. The original idea for an internal assistant started at Cisco as ChatGPT and other consumer-grade AI tools launched in late 2022 and early 2023, and Cisco’s leadership debated whether to allow employees to use them, says Srini Namineni , chief automation officer at Cisco. “The big question was, Should we actually block it?” he says. “The risks were clear when people can put company data in there, and someone else may see this data. We made a deliberate choice saying, ‘Instead of blocking it, let us give them a secure alternative.’” That initial internal AI assistant project, launched in late 2023, was also conceived to consolidate what could have become a fragmented internal AI ecosystem into one platform, while allowing employees the flexibility to connect to multiple AI models. The AI assistant, which originally included Azure OpenAI and Google Gemini, can now integrate new models within a couple of weeks of an employee’s request, Namineni says. And it has since grown into a multifunction combination copilot, coding assistant, HR assistant, and jack of all trades that allows employees to add AI capabilities to a wide range of work activities. The AI assistant, which earned Cisco a 2026 CIO 100 Award for IT innovation and leadership, saves company engineers an average of six hours per week and other employees an average of five hours, according to the company. While the main goals of the project were security and flexibility, Cisco has found a third benefit: With a monthly cost per user at about $10, the internal AI assistant operates below the subscription prices of several off-the-shelf AI assistants, Namineni says. Mitigating AI risk The tool, available to employees since 2024, had more than 96,000 users as of the first quarter of 2026, with 90% employee adoption. It has received high marks from employees, with 79% of employees believing the internal assistant saves them time, 72% saying it increases productivity, and 71% noting that it improves the quality of their work, according to internal polling.For example, the tool has accelerated software development at Cisco by helping developers find tiny bugs and generate unit tests, the company says. Namineni and his team have helped ensure employee use of the tool by continually adding new features, giving the AI assistant more functionality than some off-the-shelf assistants have. Employees can share AI prompts with one another through the assistant, and they can handle HR tasks such as scheduling vacation days without logging into another service. The assistant also enables employees to upload proprietary datasets to secure OneDrive folders for customized projects and offers retrieval augmented generation (RAG) as a service for secure querying of internal Cisco documents and metadata. The Cisco project is built on a microservices architecture, allowing for quick onboarding of new AI applications, according to the company. Cisco has pitched the assistant as an AI teammate, envisioning a virtual staff of AI agents for every employee. Evolution coming Namineni envisions several new features, including agents personalized to assist each employee on a constant basis. These personalized agents could connect to employees’ email and Webex Meetings account and take actions on their behalf, with permission. A personalized agent could sort emails based on priority, for example. He also sees the assistants taking on more HR and finance tasks, freeing up employees to perform higher-level work. Employees will have control, though, he notes. “I am not ready to let go of 100% control unless it’s a low-value activity where I’m OK with it making a mistake because it does make mistakes,” Namineni says. “Our challenge is, how do we take this power, contain it in the use cases where it can do as much work as it can, and still have human in the loop?” In addition to the CIO 100 Award, the Cisco project has earned other accolades as well. The AI assistant can be a model for other large enterprises that want to encourage employees to safely use AI, says Amy Loomis , group vice president for workplace solutions at IT analyst firm IDC. Follow the logic Other companies can adopt the logic of the approach, although they may decide not to replicate Cisco’s specific technical stack, she says. The architecture, including dual model integration, hybrid multicloud orchestration, RAG as a service, and microservices, reflects Cisco’s scale and engineering capacity, she notes. The underlying set of decisions is sound, she adds. Companies should give employees a governed internal AI environment before shadow AI proliferates; address fragmented tools with a unified intelligent interface; build access controls that keep humans accountable for what AI tools do on their behalf; and frame AI to employees as something that raises the quality and reach of their work. Cisco’s innovation lies in how the individual components of the AI assistant work as a system, Loomis says. Several individual pieces, including RAG pipelines, GPT-4o access, OneDrive integration, microservices architecture, are available elsewhere, but Cisco has put them all together. “What is less common is combining them inside a governed, enterprise-owned environment with explicit data controls, rather than routing employee queries through external AI services where sensitive data can enter public training datasets,” she adds. Loomis points to the My Projects feature that allows employees to upload proprietary datasets to secured OneDrive folders for customized Q&A. This approach gives employees the functionality they need when they otherwise would turn to unauthorized external tools, she says. Loomis also praised Cisco for framing AI as an amplification layer for human capabilities. “Giving every employee access to a set of AI tools calibrated to their role and work context is a change management choice as much as a technology one,” she says. “Organizations that position AI this way, as something that sharpens what employees can do rather than substituting for how they do it, tend to achieve broader adoption because they reduce the resistance that slows rollout.”
Score: 56🌐 MovesJul 3, 2026https://www.cio.com/article/4189683/ciscos-in-house-ai-assistant-is-a-jack-of-all-trades.html - What comes after the AI spending binge: Caps, dashboards, and the search for ROI
Uber, Microsoft, and Meta are taming runaway AI budgets. The harder question is how to measure what AI spending actually produces
- With AI costs rising, companies are hiring experts to answer a crucial question: Is it worth it?
Growing number of companies creating ‘AI enablement’ role to determine tech’s best use, quantify its effect on financial returns
Score: 55🌐 MovesJul 3, 2026https://www.theglobeandmail.com/business/article-ai-costs-rising-companies-hiring-experts-is-it-worth-it/ - How Nudge Is Reinventing Ecommerce For The Agentic AI Era
The way people discover products online is changing fast, and ecommerce brands are struggling to keep up. Instead of scrolling…
Score: 55🌐 MovesJul 3, 2026https://inc42.com/startups/how-nudge-is-reinventing-ecommerce-for-the-agentic-ai-era/ - Waymo & Uber Partnership Ends in Phoenix, While Teens Do Idiotic Things in Waymos
Waymo has partnered with Uber in a number of cities to more quickly and widely get people using its vehicles. But the partnership has ended in Phoenix. Phoenix is where Waymo first launched its robotaxi service. So, it was a bit of an odd partnership there, but the company indicates ... [continued] The post Waymo & Uber Partnership Ends in Phoenix, While Teens Do Idiotic Things in Waymos appeared first on CleanTechnica .
- Big Tech Sends Workers Into the Field to Help Customers Use AI
Microsoft, Amazon follow AI companies by creating units of “forward-deployed” engineers
- Quality Clouds Launches Hub: Enterprise-Grade Governance for AI-Generated Code
Quality Clouds Launches Hub: Enterprise-Grade Governance for AI-Generated Code Toronto Star
- Microsoft AI adoption for businesses
Microsoft AI adoption for businesses The Straits Times
- Aurora Mobile Unveils Omni-Channel AI Solutions at HKPC: GPTBots.ai Powers Enterprise Services from "Q&A" to Real Execution
Aurora Mobile Unveils Omni-Channel AI Solutions at HKPC: GPTBots.ai Powers Enterprise Services from "Q&A" to Real Execution
- Every enterprise is investing in AI; very few can prove their impact
Every enterprise now has an AI strategy, but few can prove it works. As adoption becomes universal, execution becomes the real differentiator, and independent evaluation becomes the benchmark that separates genuine impact from claimed impact: the ET Most Innovative AI Product Awards 2026 reward proof, not ambition.
- Cos keen to take bite of AI, but security and RoI hold them from sinking teeth
Industry captains say AI adoption is less about choosing latest frontier model, more about redesigning workflows and building resilience
- From tokens to trust: Sustaining, governing, and defending AI in production
Enterprises are shifting focus from AI piloting to sustainable production, emphasizing governance and defense. Key to this transition is redefining AI economics to measure useful work, not just token costs. Effective governance requires clear policies, audit trails, and intelligent workload routing. Defending AI involves robust safeguards, selective human oversight, and systems designed for potential failures, ensuring responsible scaling and risk containment for AI in real-world applications.
- Alibaba Streamlines AI Work Tools as Competition Mounts
Alibaba Streamlines AI Work Tools as Competition Mounts Caixin Global
- Robinhood's CEO says AI will trade stocks as well as humans
Robinhood has introduced two agentic products that let customers connect third-party AI agents to the platform
- AI romance scam impersonating Dubai prince ensnares victims
AI romance scam impersonating Dubai prince ensnares victims The Straits Times
Score: 54🌐 MovesJul 3, 2026https://www.straitstimes.com/world/ai-romance-scam-impersonating-dubai-prince-ensnares-victims - Running on-premise in an agentic world
On-prem AI is costly, slow, and quickly outdated versus cloud-native, continuously evolving models.
- The startups racing to put AI data centers in space before Big Tech gets there
Starcloud, Axiom Space, Lonestar, and others are betting they can stake claims in orbit before Google and SpaceX scale up
- How to conduct an AI agent security audit
My friend once raved about an AI tool he used for meeting summaries—until I asked what the tool had access to. It was only then that he realized he'd never actually looked into it. For all he knew, his AI tool could've had access to customer profiles with personally identifiable information (PII). You never know how low-stakes a tool truly is until you've investigated its connections and mapped out what it does with those connections. Here's how to conduct a security audit of your AI agent work
- Prosus flags AI and agentic AI as potential business risks as PayU turns profitable
Prosus flagged AI and agentic AI as business risks, even as PayU turned operationally profitable in FY26 and expanded payments and credit offerings across its India portfolio. The post Prosus flags AI and agentic AI as potential business risks as PayU turns profitable appeared first on MEDIANAMA .
- Unpacking Workday’s agentic AI pricing model
Only 35% of CIOs have full visibility into their AI operating costs, according to a new KPMG survey. That makes it difficult for them to control spend on software-as-a-service offerings from vendors who, like Workday, have incorporated pay-as-you-go agentic AI into their offerings. Workday is one of several vendors that have shifted to a hybrid subscription/consumption pricing model . “Fundamentally, with AI we are shifting the value of what enterprise software as a service is delivering in the industry,” Workday CTO Gabe Monroy explained in a recent interview. “The key, though, is that the value is no longer derived by a fixed factor, like how many employees you have working for you. It’s now going to be derived by how much use are you getting out of the system, hence the consumption.” However, “It’s going to be in some cases disruptive to our customers, and it’s incumbent on us to provide them with tools to forecast and navigate that transition effectively,” he said. That will be welcome news for the 40% of organizations that KPMG found have usage or token budgets in place. The changes Monroy described are part of an industry trend, according to Terra Higginson , principal research director at Info-Tech Research Group. “What we are seeing in the market is that basic seat pricing and seat counts are not going away. Customers are still paying for the core subscription footprint. AI is being layered on top as an incremental cost,” she said. “The practical message is simple: expect to pay more. The pricing model may shift from seats to credits or consumption, but the direction of spend is still up.” And because each vendor’s program has its own twists and its own ways of measuring and charging for usage, every new model adds a layer of complexity to the budgeting headaches CIOs already face thanks to the ongoing move to consumption-based services, which began with the cloud. Two parts to the model Workday’s AI pricing model is in two parts. First, customers subscribe to the services they want, as they always have. With that subscription, they receive a pool of Flex Credits that can be used to enable AI agents and other “applicable platform capabilities” including Agent-Ready Tools, Workday Data Cloud, and high-volume use of Sana through its conversational AI interface . The number of credits included varies by company size. But on top of that, they also purchase a subscription for additional Flex Credits that can be applied to any product they subscribe to. Flex Credit usage is monitored through the Platform Consumption Console, which generates alerts when consumption hits 80%, 90% and 100% of subscribed credits. Use is metered when a task is completed. However, one Flex Credit doesn’t necessarily equal one action. Workday’s rate card lists the number of credits per activity; for example, as of May 21, in the Recruiting Agent, it currently costs six credits to screen and grade each candidate’s resumé against a job opening, and 750 credits per requisition to identify relevant leads in existing talent pools and rediscover candidates for recruiters, recommending jobs for those candidates to apply for. In the Contract Negotiation Agent, the review and redlining of a contract, based on a playbook, costs 500 credits. The company also provides a reference guide listing the credits used by actions performed by the Sana platform and by self-service agents. The good news is that, though Workday’s console counts credits used in both production and pre-production environments, only those used in production are charged for, offering an early budgeting reality check and a chance to tweak processes before they land in production. Pre-production usage count is only in aggregate, however, so if a customer wants to size a specific agent, the best approach is to run it in a defined window or dedicated test tenant and compare usage before and after the test . Use them or lose them The bad news is that Flex Credits expire after one year, and any left in a subscription do not roll over to the next; it’s a use them or lose them situation. If, on the other hand, a customer exceeds their Flex Credit balance during the year, Workday said it does not just turn off their agents or other access to services. Instead, Workday’s account teams “partner with them to reconcile usage and help them purchase additional credits.” Analysts agree that there are pros and cons to this new market reality. “Workday’s Flex Credits are part of a broader shift we’re seeing across SaaS,” said Melody Brue , principal analyst at Moor Insights & Strategy. “Vendors are defining their own proprietary units for AI consumption so they can meter usage on top of existing subscriptions.” Workday’s model, she said, is more flexible than a static AI add-on because customers can use Flex Credits for whichever agents drive the most value at a given time and get access to new AI capabilities as they launch. The trade-off, however, is predictability. “Credit burn rates vary widely by task,” she said. A pilot can quietly consume a year’s worth of Flex Credits within weeks without strong telemetry and governance. And that, she said is what worries technology and finance leaders: apparently successful AI adoption that shows up as a budget surprise. But, said Scott Bickley , advisory fellow at Info-Tech Research Group, “The Workday Flex Credits Rate Card seeks to quantify consumption of Flex Credits to specific value-added actions that are AI agent-driven. Many other vendors in the ERP space have created incredibly complex, multi-layered consumption models, leaving their customers’ heads spinning as they seek to decipher how capacity will be consumed, much less if it can add value.” Brue, too, approved of Workday’s model, although she said that a core issue with AI pricing today is that vendors are each defining their own units , with no common measurement across platforms. This gives vendors pricing flexibility, but makes customers do extra work to create meaningful metrics like cost per resolution or cost per process run, just to keep budgets and ROI under control. “Workday’s Flex Credits are a smart move for Workday because they align revenue with AI usage, but from the buyer’s side, they raise the bar on FinOps and governance,” she said. “You need clear dashboards, guardrails, and forecasting, or that flexibility can quickly turn into a budget black hole.”
Score: 51🌐 MovesJul 3, 2026https://www.cio.com/article/4192838/unpacking-workdays-agentic-ai-pricing-model.html - Space is already a junkyard. AI data centers in orbit would make it worse
Low-Earth orbit is already dangerously crowded. Plans to build data centers there would accelerate a debris crisis no regulator has the power to stop
- Meta reuses old RAM in new servers with custom bridge chip
With the cost of new RAM soaring, Meta has found a thrifty way to reuse older memory in newer servers. The performance of about 40% of Meta’s millions of servers is limited by a lack of memory, the company said — but it has a surplus of older DIMMs from decommissioned servers, because RAM chips can last about twice as long as the rest of the machine. To profit from this imbalance, it developed a custom Computer Express Link (CXL) chip it calls Vistara, and associated software, to decouple older memory from server memory channels, enabling its reuse in new machines alongside their native memory. Using the older RAM with the CXL interface doesn’t significantly affect performance — although it would have done if the older DIMMs were plugged straight into newer servers. Kudos to tech site The Register for noticing the development, which Meta described in a technical paper: Vistara: Making CXL Real — Full Path from ASIC Design and OS Support to Hyperscale Deployment,” setting out how the new technology works . There is a particular need to be thrifty right now, given the current state of the market. Last year, users were warned that memory prices could double by the end of 2026, while the RAM shortage could last until 2027 . This week, Apple suggested using cheap Chinese chips , a move that may well be frowned on by the Trump administration. The Meta development may prove to be an efficient way forward. This article first appeared on Network World .
- AI skills required for 4 out of 10 graduate jobs in China, says recruitment portal
Chinese companies are seeking more fresh university graduates with backgrounds in artificial intelligence this year because widespread use of the technology is making their businesses more efficient, according to a recruitment portal based in Beijing. In the first five months of this year, nearly four out of every 10 job postings targeting fresh graduates were AI-related, compared with nearly three out of 10 in the same period last year, Maimai – a portal with 120 million users in mainland China...
- AI Factories Create Winners and Losers in Power Equipment Market
Next-generation AI factories are forcing power equipment firms to rethink their portfolios in the race to profit from a market expected to be worth more than $200 billion a year.
- The best large language models (LLMs) in 2026
Large language models (LLMs) are what most people think of when they think of AI. They're what you're interacting with in ChatGPT and Claude, they write the code generated by Claude Code and Codex, and they power other AI features like Google's AI answers and Apple Intelligence. If something has a chatbot, some kind of text generator or text summarization built in, or writes code or automates your computer, it almost certainly uses an LLM. LLMs have been studied in research labs since the late 2
- Why building AI for schools is harder than building a chatbot: inside Smartschool’s approach to exam prep
Artificial intelligence has proven that it can trawl the internet to retrieve information quickly for answering questions. But teaching students using AI is a harder task. The stakes are even higher when the goal is not just learning in school, but performing well on high-stakes exams like the SAT and ACT. On the face of […] This story continues at The Next Web
- SA’s AI lead is good news; it’s also a warning
AI changes the volume, timing and sensitivity of network traffic, which means it changes what businesses need from connectivity.
Score: 48🌐 MovesJul 3, 2026https://www.itweb.co.za/article/sas-ai-lead-is-good-news-its-also-a-warning/DZQ58MV832pvzXy2 - London’s geoSurge raises €10 million to help brands understand AI-generated outputs
geoSurge, a British DeepTech AI company helping brands shape how they are represented inside generative AI systems, announced a €10 million ($12 million) Seed round to expand its global research and engineering teams, invest in AI infrastructure and compute capacity, and accelerate development of its Corpus Engineering discipline. The round was led by AlbionVC, with […] The post London’s geoSurge raises €10 million to help brands understand AI-generated outputs appeared first on EU-Startups .
- The Rise of the “Claude Cowboy” in RevOps
A new archetype is emerging in Rev Ops: the “Claude Cowboy.” The term is gaining traction as shorthand for commercially minded operators using tools like Claude CoWork and other agentic AI tools as well as low code automation to solve operational problems fast. The Wild West Or A New Ops Utopia? Social commentary often frames […]
- AI boom lifts industrial engineers' pay packages
India's burgeoning AI data-centre sector is creating a surge in demand for mechanical, electrical, and industrial engineers. These professionals, once overlooked, are now commanding significant salaries as companies build energy-intensive AI infrastructure. Roles like cooling specialists and power procurement heads are in high demand, driving a new career path for talent in physical infrastructure development.
- From software engineering to AI engineering: Lessons from the front lines of enterprise AI deployment
"The world is shifting from software engineering to AI engineering," says Rajesh Sinha, Founder and Chairman of Fulcrum Digital — and most enterprises, he argues, are getting it wrong by treating the model as the whole solution. Drawing on hard-won lessons from deploying AI at scale inside banks, insurers, and logistics firms, Sinha makes a case that runs counter to the prevailing hype: that the real bottleneck isn't intelligence, it's infrastructure — data governance, cost control, and orchestration The post From software engineering to AI engineering: Lessons from the front lines of enterprise AI deployment appeared first on Express Computer .
- How the world's top AI models were revived
The fight that scrubbed the world's most powerful AI models from the internet featured personality clashes, industry confusion, and international backlash. Why it matters: Anthropic's models are back online, but the impact of its 20-day showdown with the Trump administration will be long lasting. Behind the scenes: It began when Amazon, Anthropic's partner and investor, sounded an alarm that was later disputed by cybersecurity experts. It warned about a "jailbreaking" issue it found with the AI lab's latest models, Mythos and Fable — meaning a technical flaw that could have caused a failure of their guardrails. Amazon flagged its concerns to the administration, triggering sweeping export controls . A U.S. official said the government conducted its own tests once it became apparent that the issue needed to be addressed. Cybersecurity experts, however, later wrote in an open letter to the administration that other leading AI models have the same issue Amazon warned about with Anthropic. On June 12, Commerce Secretary Howard Lutnick , at the direction of President Trump, called Anthropic CEO Dario Amodei. Lutnick made clear to Amodei the issue needed to be resolved fast and alerted the CEO that the company would be receiving a letter imposing sweeping export controls, the U.S. official said. Amodei called Lutnick back that night after receiving the letter, realizing it effectively meant the models would have to be taken offline — to which Lutnick responded that was indeed the goal. That decision led to a three-week, multi-agency crash course in AI safety. Anthropic deployed engineers to Washington D.C. According to a U.S. official, the company wanted to prove everything was already resolved and further changes were being fine tuned. But the federal Center for AI Standards and Innovation and the National Security Agency said those changes weren't good enough, prompting further fixes, according to the U.S. official. Gradually, various agency heads approved of the changes, and on July 1 the models were released, the official said. Out of all of the administration officials Amazon's Andy Jassy could have called, it was Treasury Secretary Scott Bessent who first heard about the jailbreaking issue found in the company report, according to a separate source familiar. Bessent was early to sound the alarm on Mythos, work with White House Chief of Staff Susie Wiles to re-engage the embattled company , and help get a cybersecurity executive order across the finish line. While technical discussions to address the jailbreaking issue took place in D.C., it was Bessent who stood next to President Trump during the G7 where allies called for global cooperation on safety standards. At the center of the showdown was Commerce Secretary Howard Lutnick , who also flanked Trump at the G7 meeting while his department's teams led technical discussions. National cyber director Sean Cairncross, the White House Office of Science and Technology Policy, Treasury Department chief information officer Sam Corcos, and the NSA also all participated in technical discussions, according to various sources. Washington mobilized faster to hold scores of meetings and pulled in far more agencies than one would expect for a single technical issue, one source said. The tension spiraled amid personality clashes and poor communication. Anthropic eventually understood that in order to be successful they needed to be on the same side as the government, the U.S. official said. As discussions turned more technical, Anthropic policy chief Sarah Heck and Anthropic co-founder Tom Brown got more involved. Brown also had multiple conversations with Lutnick and Cairncross the weekend of June 12. There was never a moment where Dario stepped offstage and someone else replaced him, one source said, adding that Brown's technical expertise allowed him to sit in a room with government specialists and go line‑by‑line through how models behave under stress. Between the lines: It remains uncertain when and how Anthropic's models will be released to ally countries around the world — which proponents say is key to beating China — or how other labs from OpenAI to Google will release their latest models. OpenAI, whose latest model GPT-5.6 is on hold, did not have visibility into discussions between Anthropic and the White House and is engaged in daily technical discussions on the release of its own model, a source said. The bottom line: There's a lot of work left to be done on a framework for approving future models with a clear inclusive process that has transparency standards and timelines, sources familiar said.
Score: 47🌐 MovesJul 3, 2026https://www.axios.com/2026/07/03/anthropic-ai-models-revived-behind-the-scenes - Tesla caps employee AI spending at $200 per week
Tesla caps employee AI spending at $200 per week, according to an internal memo reported by The Information. The article Tesla caps employee AI spending at $200 per week appeared first on The Decoder .
- Inside the fastest-growing Canadian AI startup you’ve never heard of
Ottawa-based Turbopuffer could be a blueprint for the kind of tech company Canada needs to fulfill its AI ambitions
- The 6 best autonomous AI CRM tools in 2026
My very first CRM was a DOS-based system I was forced to use in 2009. Which, just to be clear, was well after the invention of the iPhone, streaming services, and functioning graphical user interfaces. It didn't integrate with anything, it didn't suggest or automate a single thing, and if it had any "intelligence," it was the kind that treated the tab key as a threat to its authority. I've seen the CRM evolution—from command lines and black screens to the sleek, AI-infused platforms of today tha
- The future of infrastructure monitoring: From manual surveys to AI-powered geospatial solutions
By Amit Sharma, Founder and Whole Time Director at Matrix Geo Solutions Infrastructure is the physical spine that drives economic growth. Roadways, railways, bridges, dams, electrical lines, and oil and […] The post The future of infrastructure monitoring: From manual surveys to AI-powered geospatial solutions appeared first on Express Computer .
- 'It's just his AI and my AI going back and forth' The workplace phenomenon that's undermining human relationships
'It's just his AI and my AI going back and forth' The workplace phenomenon that's undermining human relationships Fortune
Score: 46🌐 MovesJul 3, 2026https://fortune.com/article/ai-communication-undermining-human-relationships-middle-management/ - The accordion effect: How AI follows the rhythm of expansion and compression
The moment we are asked to implement while still experimenting, something within starts to strain. Lately, I have been noticing this across organisations and teams working through AI integration. There is still a lot of curiosity. New tools, new ideas, new ways of working. At the same time, there is a very real shift happening […] The post The accordion effect: How AI follows the rhythm of expansion and compression appeared first on e27 .
Score: 45🌐 MovesJul 3, 2026https://e27.co/the-accordion-effect-how-ai-follows-the-rhythm-of-expansion-and-compression-20260701/ - Why job interviews are increasingly becoming AI tests
Candidates must prove how they would use tech tools in non-tech roles
Score: 45🌐 MovesJul 3, 2026https://www.irishtimes.com/business/work/2026/07/03/job-interviews-are-becoming-ai-tests/ - Meet WebBrain: An Open-Source, Local-First AI Browser Agent That Reads Pages and Automates Tasks in Chrome and Firefox
Meet WebBrain: An Open-Source, Local-First AI Browser Agent That Reads Pages and Automates Tasks in Chrome and Firefox MarkTechPost
- Pick n Pay launches AI grocery shopping assistant in South Africa
Ferigolli said more AI-powered features would be introduced in the coming months
- Your iPhone will soon warn you before you fall for a scam
Apple is introducing a fraud detection framework called Trust Insights with iOS 28 that analyzes on-device behavior to flag likely scams during calls, texts, and emails.
Score: 45🌐 MovesJul 3, 2026https://www.digitaltrends.com/phones/your-iphone-will-soon-warn-you-before-you-fall-for-a-scam/ - The National Beat: AI certifications are on the rise — and so are most paychecks
Catch up on the latest AI and startup business news in this week's Beat newsletter.