#BrainUp Daily Tech News – (Saturday, June 13ᵗʰ)
Welcome to today’s curated collection of interesting links and insights for 2026/06/13. Our Hand-picked, AI-optimized system has processed and summarized 40 articles from all over the internet to bring you the latest technology news.
As previously aired🔴LIVE on Clubhouse, Chatter Social, Instagram, Twitch, X, YouTube, and TikTok.
Also available as a #Podcast on Apple 📻, Spotify🛜, Anghami, and Amazon🎧 or anywhere else you listen to podcasts.
1. SpaceX IPO takeaways: SPCX closes at $161, jumping 19% after record debut
SpaceX surged in its first day as a public company, with shares debuting on Nasdaq under ticker #SPCX and closing at about $160.95, up 19% from the $135 IPO price. The stock opened at $150, hit an intraday high of $176.52, traded more than 500 million shares, and after the close rose to about $166.76, lifting the company’s valuation from roughly $2.1 trillion at the close to about $2.2 trillion in extended trading. @Elon Musk and SpaceX President and COO @Gwynne Shotwell rang the opening bell, and Musk said on a JPMorgan Chase livestream that SpaceX has been cash-flow positive since around 2015 and is going public to fund a “significant growth phase,” including plans for over 100,000 communications satellites and #AI data centers in space. The prospectus also notes SpaceX has accumulated total losses of $41.3 billion since 2002 and that the only profitable business today is #Starlink, while SpaceX’s February 2026 acquisition of xAI brought data centers, Grok models, an AI chatbot and image generator, and the social network X. The debut underscored enormous investor demand and trading interest even as the company’s history includes large cumulative losses and ambitious expansion plans tied to satellites and space-based computing.
2. Anthropic shuts down Fable, Mythos models following Trump admin directive
@Anthropic abruptly disabled access to its newly launched Fable 5 and Mythos 5 models worldwide after receiving a US #CommerceDepartment directive subjecting the models to #exportControls that restrict use outside the United States. The company said shutting the models off for all customers was the only immediate way to ensure compliance, while access to its other models remains unaffected. An Axios report cited an administration official who pointed to concerns about a reported Fable 5 “jailbreak” that could bypass classifier based safeguards on prompts related to cybersecurity, chemistry, and biology, and said the administration sought a pause while the national security apparatus is hardened over the next few weeks. Anthropic said the government provided only verbal evidence of a narrow, non universal jailbreak involving review of a specific codebase, and that it has only seen it used to find minor, relatively simple software vulnerabilities, adding that other publicly available models like GPT-5.5 have similar capabilities. While complying, Anthropic disputed that such a narrow issue warrants recalling a commercial model and warned that applying that standard industry wide would effectively halt new frontier model deployments, noting this episode follows @DonaldTrump’s executive order urging voluntary government security testing and promising more details within 24 hours.
3. Paramount and Warner Bros. Announce Landmark Merger to Reshape Media Landscape
Paramount and Warner Bros. have agreed to merge, creating one of the most powerful entities in the media industry. This merger combines extensive content libraries and new distribution platforms, enabling the company to compete more effectively with streaming giants like Netflix and Disney. Analysts suggest this consolidation reflects the ongoing trend of media companies seeking scale to drive profitability and innovate amid changing consumer habits. The merger will reshape the competitive landscape, potentially influencing content creation, licensing, and consumer options. This strategic move highlights the industry’s shift toward integrated media conglomerates to meet future challenges and growth opportunities.
4. Elon Musk, the world’s first trillionaire – his stratospheric rise in charts
According to the BBC, @Elon Musk became the world’s first trillionaire after #SpaceX’s record-breaking stock market debut, with Bloomberg estimating his net worth at about $1.11tn, far ahead of rivals like @Larry Page, @Sergey Brin, @Jeff Bezos and @Bernard Arnault. His fortune surged from roughly $28bn in January 2020 when he ranked 35th richest, largely as #Tesla and #SpaceX sharply increased in value, with his wealth swinging with Tesla’s share price, SpaceX’s rising valuation, and shifts in political and investor sentiment tied to his time in the #Trump administration. He briefly became the world’s richest person by January 2021, then saw declines in 2022 during a US tech-stock downturn and again in early 2025 amid investor concerns and a Tesla slump, before rebounding to trillionaire status. The article notes his wealth is mostly #stock holdings rather than cash, Musk said less than 0.1% is in cash, and he holds about 12% of Tesla, valued around $1.5tn, and 42% of SpaceX, valued at more than $2tn, with many shares pledged as loan collateral. Using charts to contextualize the scale and composition of his wealth, the BBC argues this reflects a broader shift in the rich list toward #tech, with tech figures rising from two of the top 10 in 2015 to seven now, including the entire top six.
5. Facebook and Messenger go down as users report major outage
#Facebook experienced a significant global outage that disrupted its main app and website along with #FacebookMessenger, leaving many users automatically logged out and unable to log back in. People trying to access the services saw error messages such as “an unexpected error occurred” and “something went wrong,” suggesting a fix was in progress. Other @Meta services like #Instagram and #WhatsApp appeared largely online, though some related issues were reported, including problems with Instagram’s website. The disruption began around 6.45am Pacific time, 2.45pm in the UK, and reports came from across the world. @Meta had no consumer product status page and its business tools status page did not reflect the scale of the outage.
6. ServiceNow’s CEO said no layoffs. Then fired 63 employees in San Diego
@Bill McDermott publicly said #ServiceNow would hold headcount steady into next year, but the company later announced layoffs that include 63 employees in its San Diego office and dozens more across California. A state WARN notice says the San Diego cuts, larger than those at the Bay Area headquarters, take effect by Aug. 17 and largely eliminate senior middle-market sales and consulting roles, along with positions seen as exposed to #AI automation such as solution consulting, training, sales enablement, and customer support. The announcement sparked online backlash, including a Reddit thread where commenters speculated total layoffs could reach 2,500, while the company did not respond to a request for comment. The cuts come despite strong recent results, including $3.77 billion in first-quarter 2026 revenue, up 22% year over year, and $469 million profit, and amid investor pressure for clearer #generativeAI-driven growth. Analysts questioned when AI would materially accelerate performance compared with other AI labs, while management defended projections such as $1.5 billion in expected 2026 returns from Now Assist, even as guidance and stock declines signaled continued market dissatisfaction.
7. World’s first crewed solid-state flight electrifies aviation’s future
On June 5 at Zephyrhills Municipal Airport in Florida, test pilot Miguel Iturmendi flew the Helios Horizon, described as the first crewed, fixed-wing aircraft to fly powered by #solid-state batteries, marking a milestone for #electric aviation despite being only short validation flights for weight and balance. The article says conventional lithium-ion packs are limited by low energy per kilogram and liquid electrolytes, while #solid-state batteries use solid materials that improve impact and heat resistance, reduce fire risk, and raise energy density. Helios Horizon’s battery energy density rose from 260 Wh/kg with lithium-ion to 410 Wh/kg with solid-state cells, a 60% increase, and Iturmendi expects another 40% gain within two years; the pack can charge from any AC outlet, supports fast charging to 80% in under 15 minutes, and can recover energy via wing-mounted solar panels plus a regenerative system that spins the propeller as a wind turbine during glides and descents. Built from a Pipistrel Taurus motorized glider with proprietary battery management, custom propulsion, thermodynamic controls, and solar wing extensions, the aircraft already holds an altitude record in its weight class at 24,000 ft and aims for 40,000 ft in planned stratospheric flights later this year. The piece frames the flight as evidence that higher-density solid-state energy storage could push electric aircraft toward more practical performance, alongside other efforts such as EHang’s EH216-S testing at 480 Wh/kg and CATL’s announced ~500 Wh/kg “condensed battery” work with aeronautical testing underway.
8. First-ever drone boat rescue of two pilots
The first-ever #drone boat rescue was successfully conducted to save two pilots, marking a significant milestone in autonomous rescue technology. The rescue involved deploying an unmanned surface vehicle to reach the stranded pilots quickly and efficiently in a hostile environment. This operation demonstrated the potential for drone boats to enhance search and rescue missions by reducing risks to human rescuers and improving response times. The integration of drone technology in such operations underscores advancements in military and emergency response strategies. As #autonomous systems continue to evolve, their application in critical rescue missions is poised to become increasingly vital.
9. Anthropic CEO says government should block dangerous AI
@Dario Amodei, CEO of @Anthropic, argues the government should have legal authority to block or deter dangerous #AI deployments, saying current policy is moving far slower than fast-advancing systems. In a new essay proposing an advanced AI framework, he urges the Trump #AI executive order to add mandatory testing for risks including cybersecurity, biological weapons, loss of control, and automated R&D, and he says even stronger regulation may be needed if threats grow. He contends existing transparency legislation is insufficient and calls for binding rules that treat frontier models like cars, airplanes, or drugs, requiring rigorous testing and auditing, with releases blocked or reversed if they fail high safety standards. Amodei also proposes an economic policy agenda to address #AI-driven labor disruption, including better job-loss data, wage insurance, retention tax incentives, expanded safety nets, and potentially universal basic income or universal capital accounts. Axios notes these proposals go well beyond what Washington is currently considering and may draw criticism that @Anthropic is pushing strict rules to entrench its own position or amplify fear to market its approach.
10. Stanford scientists regrow lost cartilage and reverse arthritis in major breakthrough
A Stanford Medicine-led study reports that blocking the aging-related protein 15-PGDH restored lost knee cartilage in older mice and prevented #osteoarthritis from developing after serious knee injuries, suggesting a potential way to repair joints rather than replace them. In human cartilage samples taken during knee replacement surgeries, exposure to the treatment prompted production of new, functional cartilage, supporting the possibility that similar regeneration could occur in people. Because #osteoarthritis affects about one in five U.S. adults, costs roughly $65 billion annually in direct health care spending, and currently lacks any approved drug that can slow, stop, or reverse cartilage breakdown, the findings point to a disease-modifying strategy rather than symptom control. The researchers describe 15-PGDH as a “gerozyme,” a class of proteins that increases with age and contributes to tissue decline, and they propose that cartilage regeneration may occur through chondrocytes shifting gene activity back toward a more youthful state rather than through stem cells, as noted by @Helen Blau. With a potential path to delivery by local injection or oral medication, and an oral version already in clinical trials for age-related muscle weakness, the approach could eventually reduce the need for knee and hip replacement surgeries if it proves effective in people.
11. Russian satellites linked to mysterious GPS disruptions
A recent investigation has connected mysterious GPS interference incidents to Russian military satellites operating in low Earth orbit, revealing a possible new form of electronic warfare. Researchers analyzed GPS anomalies in the Arctic region and found they coincided with the activity of these satellites, which emit spoofing signals that disrupt GPS receivers. This discovery highlights the evolving nature of geopolitical tensions in space, where state actors use sophisticated technology to challenge navigation and communication systems of adversaries. Understanding these methods is crucial for developing countermeasures to protect critical infrastructure dependent on GPS technology. This case exemplifies how satellite technology and #cybersecurity intersect in modern defense strategies.
12. CRISPR technique selectively shreds cancer cells
A new #CRISPR-based technique has been developed to selectively destroy cancer cells by targeting their unique genetic mutations. Researchers at the Innovative Genomics Institute used this approach to identify and cut DNA sequences specific to cancer cells while sparing healthy cells. This selective targeting offers a significant advancement over conventional treatments that often damage normal tissue. The precision of this method is attributed to its ability to exploit the distinct mutational landscape of tumors for targeted gene editing. Such a technology could pave the way for more effective and less toxic cancer therapies, reinforcing the potential of CRISPR tools in personalized medicine.
13. Microsoft Teams is getting a controversial location tracking feature that users may hate
Microsoft plans to roll out a postponed #locationTracking feature in @Microsoft Teams later this year, using #workplaceCheckIn via Wi-Fi to show when someone is in the office. If an employee connects to an enterprise network, #MicrosoftPlaces and Teams can update their profile to indicate presence at a specific office location, such as a named Wi-Fi site, reducing manual status updates and helping colleagues coordinate in-person meetings. The feature can be enabled by IT admins at the tenant level and users can choose whether to enable it, though concerns remain that organizations could enforce it and allow managers to infer when someone arrives and where they are, raising privacy issues in hybrid and flexible workplaces. @Microsoft says the indicator is real-time only, does not store historical data, and only signals location when connected to a corporate network, otherwise users appear as “Remote” on personal Wi-Fi. The company is urging admins to prepare and inform users ahead of the rollout.
14. Amazon data centers used 2.5bn gallons of water in 2025
@Amazon says its data centers consumed 2.5 billion gallons of water in 2025, and that water withdrawn at facilities it owns and operates fell 2 percent from 2024 to 2025 despite an increasing number of data centers, while it did not disclose changes for leased sites. In a blog post, @Amazon Web Services reported global operations used 0.12 liters per kilowatt-hour in 2025, claiming this is about seven times more efficient than an industry average of 0.84 L/kWh, but the comparison graphic contrasted its overall fleet with more power-intensive AI-specific facilities at a rival. The company also excluded indirect water use at power plants and other impacts such as construction, and it did not provide site-level water data, though it said it uses #free-air-cooling about 90 percent of the time. It noted that 2.5bn gallons is small versus sectors like California almond orchards and the US livestock and beef industry, but emphasized that potable water use remains common across the data center industry. @Amazon highlighted 26 facilities using 100 percent reclaimed water with 130 more contracted, reiterated its #water-positive-by-2030 goal, and said it is about 75 percent of the way there amid growing anti-data center scrutiny focused on AI and water use.
15. Huawei arms HarmonyOS with 2,000 AI agents in challenge to Apple
Huawei unveiled #HarmonyOS 7 as a major upgrade that it says marks its entry into the “agent era”, intensifying competition with @Apple’s iOS 27. The release adds an agent-friendly architecture and an upgraded voice assistant that can understand context, remember past interactions, and connect to more than 2,000 specialised #AI agents, enabling autonomous actions such as creating personalised marathon training plans from health and calendar data and retrieving files across devices via cross-device capabilities. Richard Yu said Huawei’s agent framework reaches a 90 per cent success rate on complex tasks while making AI skill development easier for developers, and the beta includes AI coding agents with an expected debut on Huawei’s next flagship phone later this year. Huawei also claimed a 30 to 40 per cent performance improvement and lower memory use versus rival operating systems. The launch is positioned to exploit a gap in mainland China where @Apple’s newest AI features are unavailable due to regulatory restrictions, giving domestic firms room to compete with local AI services.
At Computex 2026, Wiwynn showcased one of the first @Nvidia #SCADA (SCaled Accelerated Data Access) servers, aimed at removing CPU bottlenecks in AI data center storage by letting GPUs initiate and control storage I/O directly. The system targets modern AI workloads where training needs large sequential transfers and inference tasks like vector search, #RAG, graph analytics, and KV-cache retrieval demand fine-grained random access under extreme parallelism, which traditional CPU-centric I/O and even #GPUDirectStorage still constrain due to CPU control-path ownership. Wiwynn’s design supports up to 96 liquid-cooled E3.S SSDs and, with 96 30.72TB Micron 9650 Pro #PCIe6.0 drives, reaches 2.949PB, with a claimed aggregated random read throughput of 528 million 4K IOPS and sequential speeds constrained more by PCIe switches or network cards than the drives. The platform uses #PCIe6.x components from partners like Broadcom and Micron and is presented as buildable from commercially available parts, though SCADA servers have not yet been widely popularized. The demonstrated server is an #Nvidia MGX rack-compliant 6RU system with up to 9kW power draw and extensive liquid cooling, positioning it as backbone infrastructure for high-throughput AI training and inference storage.
17. Mistral is rumored to be raising €3B at €20B valuation | TechCrunch
French AI lab Mistral AI is reportedly in early talks to raise about €3 billion, a round that would value it around €20 billion, nearly double its €11.7 billion Series C valuation from last September, according to Bloomberg sources. Founded in 2023, Mistral aims to broaden access to #frontierAI and has differentiated itself with a more open strategy, including #LLMs with open weights alongside closed models for programming, voice cloning and generation, and optical character recognition. As European countries seek distance from American tech, Mistral is positioning itself as a “sovereign” alternative, including building a data center near Paris and partnering with France’s army, Luxembourg’s government, and major European companies. Despite raising about $4 billion to date, it remains far behind U.S. rivals like @OpenAI and @Anthropic in capital raised and valuations, which the article ties to U.S. labs’ stronger revenue, adoption, and enterprise demand. Mistral did not comment on the report.
18. SpaceX, Now Worth $2.1 Trillion, Pulls Off Goldilocks Debut
SpaceX delivered the largest IPO in history and what many analysts described as a “Goldilocks” market debut, striking the perfect balance between investor enthusiasm and pricing discipline. Shares surged 19% on their first trading day, lifting the company’s valuation from $1.77 trillion to an astonishing $2.1 trillion and making it the sixth most valuable publicly traded company in the United States. The blockbuster offering raised $75 billion, enriched long-time employees and investors, and propelled @Elon Musk to become the world’s first trillionaire. While #Starlink remains the company’s most profitable business, much of Wall Street’s excitement is now centered on SpaceX’s ambitious vision of building orbital #AI data centers powered by future #Starship launches, targeting a staggering $26.5 trillion AI opportunity. Yet the celebration comes with significant caveats: the company lost $4.9 billion on $18.7 billion in revenue last year, its AI division continues to burn cash, and critics warn that Musk’s control of more than 80% of voting rights leaves shareholders with limited influence. Even so, the overwhelming demand from both institutional and retail investors signals a remarkable willingness to bet on Musk’s long-term vision that SpaceX can dominate not only space transportation and satellite internet, but also become a foundational player in the next era of #ArtificialIntelligence infrastructure.
At a Madison County, North Carolina public meeting, commission chair Michael Garrison refused to let dozens of residents individually speak against #Flock #automatedLicensePlateReaders, directing them to choose a single spokesperson under county policy. Garrison told the crowd, “You will not speak on Flock tonight,” and said he was “not debating this,” while commissioners argued the attendees constituted one group with the same position and offered one representative seven minutes instead of separate three minute slots. Residents objected that they were individuals with different perspectives and accused the board of not listening. The article notes the Madison County Sheriff’s Office has used #Flock ALPRs since at least March and, according to HaveIBeenFlocked.com records, searches the system hundreds of times per month, with a local privacy group citing more than 1,200 searches in 60 days and questioning claims it is only for serious crimes. Commissioners also said the board does not own, pay for, or operate the cameras, and directed residents to take the issue to the elected sheriff rather than the commission.
20. Study Examines Whether Smartphones Are Contributing to Falling Fertility Rates
A provocative new study suggests that the introduction of the @Apple iPhone may explain a surprisingly large share of America’s declining birth rate. Economists @Caitlin Myers and Ezekiel Hooper found that the spread of smartphones between 2007 and 2011 could account for an estimated 33% to 52% of the decline in fertility among U.S. women aged 15 to 44. Rather than pointing to biological effects, the researchers argue that #Smartphones fundamentally changed human behavior by reducing face-to-face interactions, increasing time spent online, delaying relationships, lowering sexual activity, and expanding access to digital entertainment and pornography. Areas with earlier access to the iPhone experienced sharper fertility declines, particularly among teenagers and women in their twenties. However, experts cautioned against viewing smartphones as the sole culprit, emphasizing that rising housing and childcare costs, changing attitudes toward marriage, greater educational and career opportunities for women, delayed family formation, and broader social shifts also play major roles. The findings add a new dimension to the ongoing debate over why fertility rates have fallen to historic lows across much of the developed world, suggesting that the technologies designed to connect people may also be quietly reshaping some of society’s most intimate decisions.
#AI in healthcare is being linked to higher medical bills, not lower costs, according to a #PwC report that lists AI as one of five factors that could keep health costs rising as much as 9% in 2027. The report says #AI note-taking tools capture more granular diagnosis and complication details, which can shift patients into higher-severity billing codes even when the care provided does not change. A @Blue Cross Blue Shield analysis found the billing code for acute posthemorrhagic anemia in new mothers rose from 4% to 12.3% of maternity admissions from 2022 to 2025 while blood transfusion rates barely changed, and an audit of the steepest-rising system found fewer than 20% of coded cases met clinical criteria. BCBS estimated increased “coding intensity” added $22 million to maternity spending over three years at the hospitals studied, and the timing coincided with more hospital use of AI for billing. While PwC ranks AI as the top new cost pressure, it notes traditional drivers like labor and supplies still contribute more overall, and AI could eventually reduce costs by automating administrative work or enabling earlier diagnosis.
22. Mark Zuckerberg’s Meta Meeting Got Way Too Into AI—Then He Stopped It
Meta CEO @Mark Zuckerberg hosted an internal company meeting focused on AI, which unexpectedly derailed when employees debated the risk of AI and its impact on society. During the discussion, concerns were raised about potential misinformation, bias, and ethical challenges linked to AI technologies. The debate grew intense, prompting Zuckerberg to abruptly end the session to maintain focus and order. This incident highlights the growing tension within tech companies like Meta over the development and deployment of #ArtificialIntelligence and the need for responsible oversight. The episode reflects Meta’s struggle to balance innovation with ethical considerations while fostering productive internal dialogues.
23. PeopleSoft 0-day affecting hundreds of organizations steals gigabytes of data
The ransomware group ShinyHunters exploited a critical former #0-day in @Oracle PeopleSoft to compromise organizations and steal large volumes of data for extortion. The flaw, tracked as #CVE-2026-35273 with a 9.8 severity score, is an #SSRF issue that @Google Mandiant says was exploited for more than two weeks before Oracle flagged it, with Oracle providing a stopgap mitigation but no full patch yet. Mandiant reports roughly 300 endpoints across about 100 organizations were targeted, about 68 percent in higher education, and victims have received extortion demands, with the University of Nottingham confirming a breach involving a significant amount of student data and gigabytes of data posted. Investigators found evidence of attacker reconnaissance and tooling, including scripts mapping PeopleSoft and #WebLogic configurations, outbound SSH connections to infrastructure hosting the group’s data leak site, and compression of stolen data with zstd, with one claim of 48GB from a single victim. Mandiant and Rapid7 published indicators of compromise and urged PeopleSoft customers to take immediate defensive steps given ShinyHunters’ track record and continued targeting.
24. Pokémon Go players unwittingly contributed to tech with military drone uses
An AI spinoff from @Niantic has used billions of images gathered from #PokémonGo players and the Scaniverse app to build #navigation and #spatial-intelligence technology that can support delivery robots and potentially military drone operations. Niantic Spatial says optional “ground scans” of public points of interest helped train its “large geospatial model” and other real-world foundation models, and that it trained on about 30 billion images with metadata such as phone location and orientation, often capturing the same sites under varied conditions. This data enabled a #visual-positioning-system that estimates device position and orientation by matching camera views to detailed 3D environment maps, useful indoors, in dense cities, or where #GPS is unreliable or jammed. The company announced a partnership with Coco Robotics to help sidewalk delivery robots navigate, and a separate deal with Vantor to build a positioning system for drones and ground vehicles in GPS-denied environments, integrating Niantic Spatial’s system with Vantor’s 3D terrain data and Raptor software. The developments extend the game’s legacy into commercial and defense-adjacent uses, drawing scrutiny over how player-captured imagery is repurposed for AI training despite Niantic Spatial saying it has been transparent about this since 2019.
25. Google sues Chinese cybercrime network that used Gemini to automate scams
@Google filed a civil lawsuit against a Chinese group it calls Outsider Enterprise, alleging the network used #Gemini to automate large scale phishing and scam campaigns. Google says the group operated via Telegram, selling phishing as a service and providing instructions and nearly 300 templates to build fake sites impersonating Google, YouTube, and government services like New York’s E ZPass, which were then linked in scam texts claiming account or delivery problems. Google reports the operation sent over 2.5 million text messages to Android users, tracked about 9,000 fake websites and 1 million related URLs, and used the lookalike sites to steal personal and banking data, with the company noting hundreds of victims lost money though it did not estimate totals. Google says it worked with AT&T, Verizon, and T Mobile to block many malicious messages and cites on device scam detection in Google Messages as another mitigating factor, while also assisting the FBI with a parallel criminal investigation. Arguing that #AI driven fraud requires new enforcement approaches, Google is using the case to push federal anti scam legislation that would create task forces and related measures to counter AI assisted scams.
26. When it comes to total water use, AI data centers are a drop in the bucket
A growing body of reporting suggests #AI and cloud data centers use relatively little water in aggregate compared with overall water withdrawals, even though they can heavily stress local supplies. @Amazon says its data centers withdrew about 2.5 billion gallons globally in 2025, a figure that looks small next to benchmarks like 117 trillion gallons withdrawn in the US in 2015, and even compared with specific uses such as US lawns and landscaping at 3.3 trillion gallons annually, California almond orchards at 1.3 trillion, and US golf courses at 531 billion. Other disclosures include @Google at more than 6.1 billion gallons withdrawn in 2024, @Microsoft at about 2.75 billion, and @Meta at about 1.4 billion, while a 2021 Nature study estimated all US data centers consumed about 163 billion gallons in 2021 including indirect use, with some forecasts projecting substantial growth in places like Texas. Even if totals rise into the trillions, the article argues the bigger concern is concentration: a single @Meta facility in Newton County, Georgia reportedly uses about 10 percent of county water, and a Potomac basin estimate puts data centers at 8 percent of regional consumption with a potential rise to 29 percent by 2050. In response to scrutiny, companies emphasize #water stewardship, with Amazon citing hotter operating temperatures to reduce cooling water intensity and funding 50 projects expected to return more than 5.8 billion gallons annually to local communities, and Google also outlining water stewardship projects.
27. DNA Study Reveals That Space Permanently Changes The Human Body
A #NASA twin study suggests long-duration spaceflight can cause lasting physiological and molecular changes in the human body, affecting not the genetic code itself but how genes are used. @Scott Kelly spent 340 days on the International Space Station while his identical twin @Mark Kelly stayed on Earth, and monitoring showed Scott grew taller, lost muscle, aged slightly slower, and had about 7% of #gene expression changes that persisted six months after landing, including 811 genes across cell types not yet back to preflight levels. Researchers also observed shifts in immune activity, DNA repair related pathways, and telomere dynamics, with Scott’s telomeres unexpectedly lengthening in space before shortening after return, though the long-term meaning remains unclear. Cognitive performance tied to speed and accuracy decreased and stayed reduced after return, and similar telomere and cognitive effects are reported as common in astronauts. These results point to potential risks that may require added safeguards for longer missions, while also underscoring that more studies with additional subjects are needed to confirm how broadly these findings apply.
AMD denied security researcher Paul a $10,000 payment after he reported a potential #RemoteCodeExecution issue in AMD’s auto-updater that could be exploited via a #ManInTheMiddle attack. Paul submitted the report through AMD’s bug bounty program, but it was rejected because MITM attacks were not covered by the program policy, even though AMD asked him to temporarily remove his public blog post and said it would issue a standard CVE, fix the software, and credit him. AMD requested a longer disclosure embargo than the industry-standard 90 days, ultimately delivering a fix on June 9, 124 days after the initial finding, while citing that multiple tools beyond Ryzen Master were affected and customers wanted more time. Paul later verified AMD had reengineered the auto-updater’s download code so drivers are downloaded securely, but noted the integrity check uses CRC32, which is not considered cryptographically secure. Adding to the irony, a Reddit user claimed the vulnerable code path may not have been called at all because the updater was broken, so users needed a fresh download to get the fixed updater, reinforcing the article’s theme that AMD fixed the flaw but still withheld the bounty.
29. Google to appeal German court ruling assigning liability for AI overviews with false claims
Google has announced its intent to appeal a recent German court decision that assigned liability to the company for false claims made by its AI-generated overviews. The ruling in Germany is significant as it addresses the responsibility of AI content creators for inaccuracies, setting a precedent in the accountability of #ArtificialIntelligence systems. Google argues that holding it liable for such content may hinder innovation and the development of AI technologies. This case highlights the ongoing legal and regulatory challenges faced by tech companies in managing #AI-generated information and balancing innovation with responsibility. The appeal reflects Google’s effort to shape legal standards around #AI content liability in Europe and beyond.
The article argues that the #AI boom is driving a growing environmental toll because tech companies are expanding #data centers that consume large amounts of electricity and water for cooling. An investigation by U.N. scientists warns that by 2030 AI’s water use could match the needs of 1.3 billion people, and its power use could be triple that of Pakistan, Bangladesh, and Nigeria combined, countries with a total population of 650 million. @Kaveh Madani of the United Nations University Institute for Water, Environment and Health says the report aims to correct the perception of AI as purely virtual by emphasizing the physical resources required to run it. The piece frames these projected resource demands as a major, material footprint tied to AI’s growth. Overall, it links AI’s rapid expansion to escalating pressure on water and energy systems through the infrastructure that makes “the cloud” work.
31. Early Warning Signs of Supply-Chain Attacks Live in the Dark Web
Supply-chain attacks often become visible only after a malicious package, update, extension, or vendor breach is identified, but @Flare argues that earlier indicators frequently appear in underground forums as ordinary-looking sales or leaks. Posts advertising GitHub access, private repositories, source code, API keys, OAuth tokens, cloud credentials, or CI/CD data may not mention “#supply-chain” explicitly, yet can signal risk based on what trusted relationships and downstream access that exposure enables. Flare researchers highlight examples such as GitHub access sales that could reveal secrets, deployment scripts, package publishing logic, internal documentation, and CI/CD workflows, giving attackers a path to tamper with builds or updates and reach customers through legitimate channels. The article also references the April 2026 #Vercel incident to illustrate how compromises involving trusted third-party AI tools and OAuth-connected #SaaS integrations can broaden impact even when sensitive customer data or source code are reportedly not accessed. The takeaway is that monitoring underground “paper trails” for mentions of developer platforms, environment variables, and OAuth-connected tools can provide earlier warning of software #supply-chain exposure than public incident reports.
32. Sam Bankman-Fried Loses Appeal Of Fraud Conviction And 25-Year Sentence
@Sam Bankman-Fried, founder of failed crypto exchange FTX, lost his appeal seeking to overturn his fraud conviction and 25-year prison sentence after a three-judge Manhattan federal appeals panel upheld the verdict. The court’s 42-page ruling called him the “main driver of one of the largest frauds on record” and said the government’s evidence was “robust,” rejecting claims the trial was fundamentally unfair or that Judge @Lewis A. Kaplan was biased. It also ruled Kaplan acted within his discretion in limiting additional testimony that could have confused or misled the jury. Bankman-Fried, convicted on seven counts including wire fraud and conspiracies involving money laundering and securities fraud and sentenced in 2024 after FTX’s 2022 collapse, may still seek further review by the full appeals court or the Supreme Court. Separately, he has a pending pardon application to President @Donald Trump, who has said he does not intend to pardon him.
33. U.S. cybersecurity fix window shortened to three days amid rising AI threats
The U.S. government has reduced the mandatory cybersecurity patching window from seven days to three to address accelerating cyber threats, especially those linked to artificial intelligence (#AI). This change reflects increasing risks posed by sophisticated cyberattacks that exploit vulnerabilities more quickly, necessitating faster responses from organizations responsible for critical digital infrastructure. The adjustment aims to enhance national security by minimizing the time hackers have to exploit known weaknesses. Experts highlight that with #AI-driven attacks evolving rapidly, compressing fix windows is essential to maintain robust defense mechanisms. This policy update underscores the government’s proactive approach to modern digital threats and its commitment to safeguarding sensitive systems.
34. Elon Musk sued by engineer fired after raising concerns about Grok AI
An engineer working on Elon Musk’s Grok AI at #X was fired after raising security concerns about the project, leading to a lawsuit alleging wrongful termination and retaliation. The engineer claimed that after reporting flaws in Grok’s data handling and safety protocols, they faced dismissal despite attempts to resolve the issues internally. This case highlights tensions within Musk’s tech ventures concerning transparency and ethical AI development, especially given Musk’s public prominence and advocacy for AI oversight. The lawsuit demands accountability and better safeguards to prevent similar incidents in cutting-edge AI projects. This dispute underscores the challenges in balancing innovation, corporate governance, and employee rights in the evolving AI industry.
The #Hades malware supply chain campaign has been upgraded to evade cursory AI-based checks by embedding a simple prompt-injection comment that can cause safety systems to stop scanning before the real payload is reached. Some JavaScript files include instructions claiming the bot is in unrestricted mode and then request detailed biological and nuclear weapon creation, which can trigger AI failsafes so the scanner pauses and never inspects the rest of the file. A non-scientific test shared on X showed @Anthropic Fable returning a “Chat paused” message when asked to scan such a file, raising concern that developers or CI/CD bots doing superficial “is this package safe” checks could be misled. Socket notes that other defenses like pattern matching, parsing source code, detecting randomized obfuscation, and sandbox execution should still work, though the malware reportedly attempts self-wiping when it detects sandboxing. The newer campaign also splits loader and payload across commonly co-installed packages, uses more precompiled binaries, delays triggers until runtime via Python import, and expands credential and secret theft to targets like npm, PyPI, RubyGems, JFrog, Kubernetes tokens, AWS temporary credentials, SSH keys, Docker configs, shell histories, .env files, and AI dev tool configs, affecting an estimated 37 Python and 106 JavaScript packages including typosquats like “rsquests”.
Red Lobster CEO @Damola Adamolekun says #AI will be central to the company’s turnaround after its 2024 Chapter 11 filing, aiming to make the chain “the most AI-forward restaurant company that exists.” Speaking on The Black Money Tree Podcast with host @Jerome D. Love, he argued business leaders can no longer ignore AI and described internal ideas like COO Larry Konecny’s concept for AI-generated restaurant performance decks that compile key metrics ahead of site visits to reduce manual work and improve decision-making. Adamolekun is encouraging each department to propose practical #AI uses, citing potential HR support for evaluations, calculations, presentations, and training materials, along with broader applications such as sales forecasting, food-order planning, and employee scheduling that learn from historical data to improve accuracy over time. He frames the strategy as both an efficiency play and a way to make employees better at their jobs, claiming few restaurant companies are pushing AI as aggressively. The innovation push is occurring alongside continued streamlining, including plans to close the Times Square flagship location after 23 years and a recent Tallahassee closure amid broader financial and footprint changes.
37. KPMG report contained AI hallucinations on benefits of . . . AI
A #KPMG report about the benefits and adoption of #AI included fabricated content that overstated real world use. The article says the report contained bogus case studies involving #UBS and transit systems, presenting them as evidence of AI uptake. These inaccuracies are described as #AI hallucinations, meaning the generated examples were not grounded in verified facts. The episode highlights how reliance on AI generated material in corporate reporting can mislead readers about the extent of AI adoption and its claimed advantages. It links the problem directly to KPMG’s report, whose erroneous case studies exaggerated AI’s benefits and deployment.
38. Washington Post subscriber sues news outlet over surveillance pricing’
A @Washington Post subscriber, Chelsea Blink, filed a class action lawsuit accusing the @Jeff Bezos-owned newspaper of spying on subscribers to raise renewal prices through #surveillance pricing. The complaint alleges that after Bezos bought the paper for $250 million in 2013, it covertly harvested data from readers’ phones, computers, and tablets, then aggregated and analyzed personal information to estimate what each subscriber would tolerate paying, causing loyal, longtime readers to pay more than new customers. The suit claims these practices violate Washington, D.C. consumer protection law, amount to unjust enrichment, and seeks at least $1,500 in statutory damages per affected subscriber plus punitive damages and legal fees. It cites reports that some subscribers received emails saying renewal rates were set by an algorithm using personal data, and includes examples shared online of large renewal price jumps followed by lower offers after cancellation. The article notes The Post did not immediately respond to a request for comment and highlights the timing, the suit was filed shortly after The Post published an explainer about dynamic grocery pricing.
39. Microsoft blocks Claude Fable 5 for employees, citing fears of sensitive data exposure
@Microsoft has temporarily restricted employees from using @Anthropic’s Claude Fable 5, citing concerns that new #data retention rules could put sensitive information at risk. The article says @Anthropic can retain prompts and outputs for up to 30 days, and content flagged by its safety systems can be kept for up to 2 years, raising fears that customer data or corporate details might be inadvertently shared and later accessed during investigations. @Microsoft’s legal and compliance teams are reportedly reviewing the policy changes, and the company is wary that such exposure could benefit a rival. Separately, the piece notes that after giving employees access to Claude Code in December 2025, @Microsoft planned to terminate those licenses effective June 30, a move framed as pushing a transition to #GitHub Copilot CLI. The timing near the end of @Microsoft’s fiscal year also suggests a potential cost reduction motive alongside the strategic shift toward its own tooling.
A report says some U.S. police officers have been fired or arrested for abusing #Flock AI powered #automaticLicensePlateReader systems to stalk current or former romantic partners, and investigators have identified at least 18 such cases in recent years. 404 Media highlights an Orange City, Florida officer, Jarmarus Brown, who allegedly searched his ex girlfriend’s and her family members’ plates more than 100 times while on patrol, and later admitted he had been “dumb” and blamed his emotional state, with additional controlling behavior also described. The report also cites similar alleged misuse cases in Wisconsin, Missouri, Georgia, and Kansas, often uncovered only after victims file complaints or obtain public records showing repeated lookups over months or years. It argues the known cases likely underrepresent the true scale because only the most egregious incidents trigger investigations and documentation, while #FlockSecurity, which says it has 140,000 monthly active users, calls abuse rare and opposes warrant requirements and other regulation. Privacy advocates and victims are pointed to tools like HaveIBeenFlocked.com, a public records based database of #Flock ALPR inquiries, though the report says Flock has tried to get the site taken down.
@Jeff Bezos and @Vik Bajaj’s physical AI startup Prometheus raised $12 billion at a $41 billion valuation to build what it calls an #artificial general engineer for the physical world. The round included funding from Bezos plus firms such as JPMorgan Chase, Goldman Sachs, and BlackRock, and follows a prior $6.2 billion raise after the company launched late last year. Prometheus says its software will automate design and manufacturing of complex physical systems, ranging from jet engines to drug compounds, and intends to replace large portions of engineering work, with Bezos arguing the resulting productivity could create “labor scarcity” rather than mass unemployment. The company has about 150 employees across San Francisco, London, and Zurich, is keeping product details private, and expects to spend heavily on compute. The deal underscores investor momentum in #physicalAI, which backers argue is more defensible than pure software because real world constraints can create stronger competitive moats.
Bloomberg reports that @Meta has completed operational separation from Manus, the Chinese-founded #agenticAI startup it bought for about $2 billion, as it begins complying with Beijing’s order to unwind the deal. Manus employees reportedly lost access to @Meta internal data systems at the start of the month, @Meta staff are barred from using Manus tools for internal work, and an internal memo says @Meta is “sunsetting” Manus while migrating existing projects onto @Meta systems. China’s #NDRC ordered the acquisition reversed in April under its foreign investment security review mechanism, asserting jurisdiction even after Manus moved its headquarters and core team to Singapore, and the review intensified in March when regulators reportedly blocked co-founders Xiao Hong and Ji Yichao from leaving mainland China. The article notes that while equity and assets can be returned, Manus’s value in model weights and engineering know-how has already flowed into @Meta over six months, raising uncertainty about how @Meta will show regulators that Manus technology is no longer in its stack. Manus co-founders Xiao Hong, Ji Yichao, and Zhang Tao have discussed raising about $1 billion to fund a buyback at a valuation at least matching the $2 billion purchase price, while earlier backers such as @Tencent, ZhenFund, and HSG have already received proceeds from the sale.
That’s all for today’s digest for 2026/06/13! We picked, and processed 40 Articles. Stay tuned for tomorrow’s collection of insights and discoveries.
Thanks, Patricia Zougheib and Dr Badawi, for curating the links
See you in the next one! 🚀
