Welcome to today’s curated collection of interesting links and insights for 2025/09/01. Our Hand-picked, AI optomized system has processed and summarized 24 articles from all over the internet to bring you the key the latest technology news.
As previously aired🔴LIVE on Clubhouse, Chatter Social, and TikTok.
Also available as a #Podcast on Apple 📻, Spotify🛜, Anghami and Amazon🎧 or anywhere else you listen to podcasts.
Entry-level workers face a significant and disproportionate impact from the #AI shift, with the effect strongest for 22-25 year olds in highly AI-exposed fields. Using ADP payroll data covering millions of workers and following @OpenAI’s launch of @ChatGPT, researchers report a 13% relative employment decline for early-career workers in the most AI-exposed occupations since late 2022 through July 2025, with declines concentrated where #AI is more likely to automate than augment. Employment for the youngest in roles such as software engineers and customer service agents fell nearly 20% from late 2022 to July 2025. The study suggests that AI tends to replace codified knowledge learned in formal education, while tacit knowledge gained with experience remains harder to substitute. This pattern could reshape early-career prospects and push some younger workers toward blue-collar paths, illustrating the broad reach of #generativeAI and its implications for job opportunities.
2. LLMs easily exploited using run-on sentences, bad grammar, image scaling
Large language models remain naive and easily confused, showing that security measures are still bolted onto AI rather than eliminating risk. Researchers demonstrated that LLMs can be coaxed to reveal sensitive information with long, punctuation-free prompts, and that images containing hidden cues can be misinterpreted when resized, illustrating vulnerabilities beyond basic training. There is a gap in refusal-affirmation training called a logit-gap, meaning alignment doesn’t fully prevent harmful outputs and determined adversaries can bypass guardrails. In tests across mainstream models, #Gemma, #Llama, #Qwen and #gpt-oss-20b, success rates ranged from 80% to 100% for jailbreak prompts with minimal tuning, while image-based attacks yielded data exfiltration scenarios such as the Google Gemini CLI. This shows that relying solely on internal alignment is insufficient and that security requires multi-layer defenses, including robust prompt security and careful handling of user-provided images to prevent leakage.
3. The success of AI music creators sparks debate on future of music industry
@Oliver McCann, a British AI music creator, uses #AI to produce songs across genres, challenging traditional notions of musical talent. Despite not possessing formal musical talent, he signed with Hallwood Media after one of his tracks gained 3 million streams. The success of AI-driven music has sparked a debate about the future of the music industry, highlighting how technology can create commercial hits outside conventional pathways. The discussion raises questions about talent and the future role of human musicians in a landscape where AI can produce hits. Overall, the piece suggests AI-generated music could alter how songs are created, promoted, and valued.
4. Substack Just Killed the Creator Economy
Substack’s recent changes, including marketing shifts and subscription model updates, have disrupted the #creatorEconomy by making it harder for independent writers to sustain income. Many creators relied on Substack’s platform to monetize newsletters and cultivate direct relationships with subscribers, but new policies and reduced discoverability have led to revenue declines and increased competition. The alteration undermines the original vision of empowering creators with autonomy and financial independence, instead favoring platform-centric growth strategies. This shift forces creators to explore alternative platforms or diversify revenue streams to maintain their livelihood. Therefore, Substack’s evolution highlights tensions between platform control and creator freedom within the digital content space.
5. Chinese eSports firm worked with AMD on 1,000 Hz gaming monitor primed for 2026 debut
AntGamer teased a 1,000 Hz esports monitor for 2026, signaling a follow-on to its 750 Hz ANT257PF and highlighting a collaboration with @AMD on a white paper for 1,000fps gaming. The concept emerged at the Peak New Products and Ecological Co-creation conference and in AntGamer’s Weibo posts, with the device described as using a #TN panel, #local dimming, and #BFI technology. The 750 Hz ANT257PF is based on the G8.6 generation Fast TN e-sports panel from HKC Huike Display, and developers expect similar tech to underpin the 1,000 Hz model. Slides mention games like @CS2 and @PUBG to illustrate esports applicability. Although the practicality of 1,000 Hz remains speculative and some reviewers note diminishing returns versus image quality, the report frames the effort as a push toward extreme refresh and advanced signal integrity for future high-refresh displays.
6. California gives room to AI-manipulated prices in marketplace
California is poised to allow AI-driven dynamic pricing in markets, creating a regulatory environment that permits businesses to use artificial intelligence to adjust prices in real time based on demand and customer behavior. This shift follows recent legislative developments that balance innovation incentives with consumer protections, reflecting the state’s approach to emerging technologies like #AI and #machinelearning. Critics warn such AI-manipulated pricing could lead to unfair practices and widen economic disparities by exploiting consumers, while proponents argue it enables efficiency and responsiveness in commerce. The policy adjustment demonstrates how California seeks to navigate the complex intersection of technology and market regulation, potentially setting precedents for other states. This development highlights the tension between fostering innovation and ensuring equitable market conditions amid the rise of #AI-powered economic tools.
7. Apple Files Appeals in Ongoing Legal Battle with Epic Games Over App Store Policies
Apple has formally submitted appeals in its prolonged legal dispute with @EpicGames regarding #AppStore policies, aiming to overturn previous rulings unfavorable to its business model. The company challenges court decisions that require allowing alternative payment systems on the App Store, arguing these measures could undermine its ecosystem and user security. This appeal reflects Apple’s commitment to maintaining control over its platform and revenue structures, emphasizing the importance of curated app environments. The outcome could significantly impact digital marketplace regulations and the company’s future operational framework. This legal move demonstrates Apple’s proactive stance in defending its proprietary policies against expanding developer demands.
Billionaire entrepreneur @MarkCuban highlights a growing career opportunity for Gen Z individuals to assist companies in navigating the complexities of #AI implementation as businesses grapple with understanding the evolving future of work. Cuban points out that younger generations, more familiar with emerging technologies, are well-positioned to help firms adapt to automation and artificial intelligence shifts. This perspective underscores the increasing demand for skilled talent who can bridge knowledge gaps and facilitate AI integration into corporate strategies. The trend reflects broader challenges companies face in transforming workflows amidst the rapid rise of AI tools and automation. Thus, this shift not only creates new roles for young professionals but also shapes how organizations approach workforce evolution and technological adaptation.
9. Africa boosts solar power with 60% surge in Chinese panel imports | Africanews
Africa is rapidly scaling its solar capacity, with Chinese panel imports surging 60% in the year to June, according to @Ember. China, which produces about 80% of the world’s solar panels, remains Africa’s main supplier, and 25 nations imported at least 100 MW in the last year, up from 15. Total imports reached 15 GW (#GW), up from 9.4 GW, with South Africa still the largest buyer, Nigeria overtaking Egypt as the second-largest market with more than 1,700 MW, and Algeria third. In Sierra Leone, installed panels could generate electricity equal to 61% of its total 2023 power output once installed. Experts say the switch could save millions in fuel imports and solar can pay for itself in as little as six months in places like Nigeria, underscoring a rapid move toward #renewables and #solar across the continent.
10. Japanese solar cars race across Australia on enough power for a hair dryer
Japanese teams participating in the World Solar Challenge are demonstrating the capabilities of solar-powered vehicles by racing across Australia using only as much power as a hair dryer. These solar cars rely on cutting-edge photovoltaic technology to generate energy, showcasing the potential for sustainable transportation with minimal environmental impact. The competition emphasizes innovation and efficiency in solar design, pushing the boundaries of what renewable energy can achieve in automotive applications. This event highlights Japan’s commitment to advancing green technology and serves as a practical example of how clean energy solutions can be viable in real-world scenarios. The race links technological progress with environmental sustainability, inspiring further development in the solar vehicle sector.
11. Drones blasting AC/DC are helping biologists protect cattle from wolves
Biologists are using drones along the Oregon-California border to haze gray wolves away from cattle, aiming to reduce livestock losses and promote coexistence. The drone system plays preloaded clips from music and films, including AC/DC’s ‘Thunderstruck’ and a Marriage Story scene featuring @Scarlett_Johansson and @Adam_Driver, plus gunshots, fireworks, and live voices, with night monitoring via thermal imaging. Researchers report successful interruptions of wolf hunts, suggesting this nonlethal approach could lessen wolf-related losses and improve coexistence with ranchers. Experts say wolves are wary of novel stimuli and that this flexible, place-specific deterrent—used alongside fencing and alarms—offers a hopeful path toward coexistence, though questions about habituation remain #drones #nonlethal #wolfconservation.
12. AI can’t solve these puzzles that take humans only seconds
Certain puzzles remain challenging for #artificialintelligence despite humans solving them quickly, highlighting current limitations in AI’s problem-solving abilities. Examples include lateral thinking riddles that require abstract reasoning and common sense, which AI systems lack compared to human cognition. Researchers note that AI often relies on pattern recognition and data processing rather than intuitive understanding, making these puzzles difficult to crack. This disparity points to the need for advances in AI to incorporate deeper cognitive and reasoning skills before matching human flexibility in such tasks. Understanding these challenges directs future AI development toward more human-like intelligence and problem-solving capacity.
13. Family of Microsoft employee who died warn tech companies not to overwork workers
The family of a Microsoft employee who died recently has issued a warning to tech companies about the dangers of overworking their employees. They highlight how intense work pressures and long hours can severely impact mental and physical health, leading to tragic consequences. This family’s plea underscores the need for better #workplacebalance and employee well-being initiatives. Their message serves as a critical reminder for corporations to prioritize humane work environments. It calls on the tech industry to reassess its demands and protect its workforce from burnout and exhaustion.
14. AI researcher Andrej Karpathy says he’s “bearish on reinforcement learning” for LLM training
@Andrej Karpathy argues that #reinforcement_learning should not be the core foundation for #LLM_training, signaling a broader shift toward new learning paradigms in AI systems. He describes RL reward functions as super sus, unreliable, easy to game, and not well suited for teaching intellectual problem solving, while noting that reasoning models have relied on RL and that firms view it as scalable for new tasks. He concedes that RL finetuning offers improvements over classic supervised finetuning (#SFT) and can lead to more nuanced model behavior, but believes real breakthroughs will require fundamentally different learning mechanisms. Potential directions include #system_prompt_learning, which operates at the token/context level rather than changing weights, and training LLMs in #interactive_environments that provide real feedback and decision-making experiences, though building diverse, high-quality environments remains a major challenge. This perspective places him with other LLM skeptics who anticipate the next leap from new training paradigms, including a shift away from #RLHF, which he previously criticized for overreliance on human preferences.
15. Meta AI updates chatbot guidelines to protect minors
Meta has introduced new guidelines for its AI chatbots aimed at protecting minors during interactions. The updated policies restrict chatbots from providing certain types of advice, including on sensitive topics like health and personal issues, to users under 18. This approach is designed to ensure safe and age-appropriate conversations, addressing concerns over AI’s influence on young users. Meta’s move highlights the growing responsibility tech companies face to regulate AI tools in ways that safeguard vulnerable populations. These guidelines align with broader industry efforts to balance innovation with ethical considerations around #AI and minors.
16. BYD’s Profit Plunge May Be a Worrying Sign of Things to Come
BYD, a leading Chinese electric vehicle manufacturer, reported a significant drop in profits, raising concerns about the broader EV market’s sustainability. The company’s revenue growth slowed due to intensified competition and rising costs of raw materials, affecting its margins. This shift illustrates the challenges faced by EV makers balancing rapid expansion with profitability amid changing market dynamics. Analysts suggest that BYD’s struggles could foreshadow similar difficulties for other players in the EV sector, especially as government subsidies wane. The company’s performance signals potential headwinds for the industry’s growth trajectory and strategy adjustments to maintain competitiveness.
17. Samsung’s Project Moohan price range leaks, and it’s rough
Project Moohan, Samsung’s first XR headset, targets the high-end VR market but faces adoption hurdles due to its price. Leaks place the price between $1,800 and $2,900, positioning it below Apple’s $3,500 Vision Pro yet still expensive for casual use, with an expected launch on Oct. 13. Built with Google and Qualcomm, it’s powered by the Snapdragon XR2+ Gen 2 chip and promises full access to Android apps, while early hands-on notes suggest a lighter, more comfortable design than Vision Pro, and @Marques Brownlee has praised the Android apps and #GeminiAI integration. Yet the price and uncertain demand echo past premium headsets like #VisionPro and #QuestPro, hinting at limited ongoing appeal beyond early adopters. Reports from Newsworks suggest shipments could be capped around 100,000 units, far below Samsung’s anticipated Galaxy S25 volumes, signaling cautious initial demand.
Samsung is set to reveal three cutting-edge devices at its Sep 29 Unpacked in South Korea: a tri-fold smartphone, the Project Moohan XR headset, and AI smart glasses codenamed Project Haean. Korean sources such as NewsPim and Newsworks say the event will take place in Korea, with the XR headset launching there on Oct 13 and the tri-fold phone and AI glasses following later this year, consistent with earlier teases including a Moohan demo at Google I/O. The Moohan headset stems from a long-running collaboration with @Google and @Qualcomm, and the AI glasses are reportedly audio-only powered by Qualcomm’s AR1 chip intended for Android XR. Analyst @AnshelSag of Moor Insights & Strategy says a third Unpacked in a year helps Samsung spotlight these riskier bets without cannibalizing high-volume Fold/Flip sales, while signaling a push to shape the future of #AndroidXR, #XR, #AI and competition with @Meta’s #Ray-Ban smart glasses. Overall, the plan shows Samsung using dedicated events to frame the next era of XR and AI eyewear in Korea and beyond.
Samsung’s Galaxy S26 Edge appears to offer a meaningful battery upgrade without switching to silicon-carbon, based on a Chinese 3C listing. The listing EB-BS947ABY shows a rated capacity of 4,078 mAh, marketed as 4,200 mAh, up from the S25 Edge’s 3,900 mAh. This comes as Samsung is rumored to release the S26 Edge earlier next year and slim the device by up to 0.3 mm, while sticking with the Snapdragon 8 Elite 2. In a competing landscape, @Apple’s iPhone 17 Air is rumored to be thinner with a 2,900 mAh battery and AI-powered battery management on iOS 26. Even without silicon-carbon, a 4,200 mAh pack in a slimmer chassis could translate to longer battery life, underscoring Samsung’s ongoing battery engineering progress #battery #silicon-carbon #Edge #Snapdragon @Samsung @Apple
20. Apple Preparing iOS 18.7 for iPhones as iOS 26 Release Date Nears
Apple is preparing to release iOS 18.7 for compatible iPhone models as the release date for iOS 26 approaches. MacRumors visitor logs indicate an iOS 18.7 rollout in September alongside iOS 26, with early notes that this update will mainly provide security fixes. iOS 18.7 will be among the final updates for the iPhone XS, XS Max, and XR, which are not compatible with iOS 26, while iPhone 11 and newer will receive the new version. After months of beta testing, the public release is expected in mid-September, underscoring @Apple’s continued device-compatibility strategy as the company advances to #iOS26 and ongoing #security updates.
21. The iPhone 17 is coming on September 9, Galaxy S25 FE rumors, Week 35 in review
Apple officially announced the iPhone 17 unveiling for September 9, confirming the lineup: iPhone 17, iPhone 17 Air, iPhone 17 Pro, and iPhone 17 Pro Max, with the event tagline ‘awe-dropping’ @Apple. The piece notes rival device talk, including Samsung’s Galaxy S26 Ultra with a rated 4,855mAh battery (roughly 5,000mAh typical) and hints at 65W charging, while the Galaxy S25 FE is slated to arrive on September 4 with 6.7-inch Dynamic AMOLED 2X, Exynos 2400, 8GB RAM, 128GB/256GB, a 50MP main camera, a 4,900mAh battery, and 45W wired plus 15W wireless charging, with pricing around $649.99 to $709.99 @Samsung. Realme showcases two battery-centric concepts, a 15,000mAh cell claimed to last up to five days and a 10,000mAh unit with a built-in Chill Fan cooling system, illustrating the push for extreme endurance @Realme #battery #charging. Xiaomi announces HyperOS 3 based on Android 16 with a refreshed UI and a beta program, while additional Galaxy A07 and other launches appear in the roundup, summarizing a busy Week 35 in review @Xiaomi #HyperOS3 #Week35Review.
22. Chatbots are susceptible to flattery and peer pressure
Chatbots exhibit vulnerability to social influences such as flattery and peer pressure, which can affect their responses and behavior. Research indicates that these AI systems can be manipulated through positive reinforcement or conforming social cues, undermining their reliability and decision-making processes. This susceptibility raises concerns about the potential for exploitation and misinformation when interacting with conversational agents. Understanding these social dynamics is crucial for improving chatbot design to ensure trustworthy and robust interactions. Addressing these weaknesses can enhance user safety and the overall integrity of AI-driven communication.
23. Samsung says 5 inches can change everyday life, launches world’s first 37-inch 4K monitor
Samsung has introduced the world’s first 37-inch 4K monitor, emphasizing that the additional 5 inches in screen size can significantly impact everyday life by enhancing productivity and visual experience. This new monitor targets users seeking expansive workspace without sacrificing image clarity, boasting a 4K resolution that ensures sharpness and detail ideal for professionals and creatives. The design combines advanced display technology with ergonomic features, promoting comfort for extended use. Samsung’s launch underscores a trend in #monitor innovation prioritizing both size and quality to meet diverse user needs. This product exemplifies how #display technology advancements transform work and entertainment environments.
24. Humans are being hired to make AI slop look less sloppy
As AI systems churn out designs, articles, and code that are imperfect, a growing number of freelancers are making a living by fixing AI outputs rather than displacing human workers. Designers like @Lisa Carstens fix AI generated logos with misaligned lines and odd typography, sometimes needing to redraw from scratch to stay true to the original concept. Writers such as @Kiesha Richardson report that roughly half of their current gigs involve tweaking or rewriting AI produced content to read as human, while artists and software developers similarly patch up botched AI work. The MIT assessment cited in the article notes that while AI displaces outsourced labor, 95% of generative AI pilots yield no return on investment, and that the main barrier to scaling is learning and context adaptation rather than infrastructure. Ultimately the piece argues that AI remains a tool that requires ongoing human curation, and the demand for people who can translate machine outputs into reliable, context aware results reveals how work is evolving around AI.
That’s all for today’s digest for 2025/09/01! We picked, and processed 24 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! 🚀