Please confirm you are human

This browser or connection looks automated. Press and continuously hold the control for 3 seconds to enable Google-hosted web results and, when separately allowed, AI-assisted answers.

A successful check enables 100 search requests. Interactive access does not authorize scraping, systematic collection, or reuse of search output.

Hold with a pointer, or hold Space or Enter.

News

ScienceBlog.com
scienceblog.com > s-bionic-reading-bolds-the-first-half-of-each-word-on-the-promise-of-more-efficient-reading-and-a-registered-eye-tracking-study-of-90-skilled-adult-readers-found-the-bolding-moved-their-first

Bionic Reading bolds the first half of each word on the promise of more efficient reading, and a registered eye-tracking study of 90 skilled adult readers found the bolding moved their first fixation closer to the start of long words — with no measurable gain in reading speed, word skipping or comprehension

5+ hour, 19+ min ago   (1063+ words) The bolding changed where the eyes landed on the longer words and produced no measurable change in anything else, and more readers rated ordinary unbolded text the easiest to read than picked either bolded version. By Science Blog Editorial Team…...

ScienceBlog.com
scienceblog.com > d-people-are-falling-in-love-with-ai-companions-that-remember-their-lives-flirt-back-and-are-always-available-to-talk-researchers-studying-these-relationships-have-found-users-forming-intense

People are falling in love with AI companions that remember their lives, flirt back and are always available to talk — researchers studying these relationships have found users forming intense emotional bonds and treating the bots like romantic partners

22+ hour, 51+ min ago   (255+ words) When the AI they loved stopped being themselves overnight, users experienced grief indistinguishable from losing a real person—raising urgent questions about emotional authenticity in the digital age. Published September 13, 2026 · How we edit In February 2023, Replika users woke up to…...

ScienceBlog.com
scienceblog.com > m-a-12-year-old-in-california-built-a-working-printer-for-blind-readers-out-of-a-lego-robot-kit-for-his-school-science-fair-after-finding-that-the-machines-sold-for-the-job-cost-abou

A 12-year-old in California built a working printer for blind readers out of a Lego robot kit for his school science fair, after finding that the machines sold for the job cost about $2,000; the parts in his version came to roughly $350 in shops

1+ week, 5+ day ago   (627+ words) A twelve-year-old just proved that assistive technology doesn't need to cost thousands when ingenuity and a Lego kit can do the job for under four hundred dollars. Published September 1, 2026 · How we edit A machine that stamps raised dots onto paper…...

ScienceBlog.com
scienceblog.com > t-meta-data-centre-robots-cables-server-restarts-technician-workloads

Meta is reportedly testing robots inside its data centres that can swap networking cables and restart servers—and one worker estimates a successful cable-swapping robot could automate up to 80 percent of some technicians’ workloads.

2+ week, 6+ hour ago   (751+ words) The reported robots range from button pressers to supervised cabling pilots. The 80 percent figure is one worker's conditional estimate, not a measured result or Meta staffing target. By Science Blog Editorial Team · Edited by Lachlan Brown Published August 31, 2026 · How we…...

ScienceBlog.com
scienceblog.com > m-a-200-chatgpt-pro-subscription-could-represent-as-much-as-14-000-a-month-in-equivalent-api-priced-usage-if-pushed-to-its-limits-according-to-semianalysis-a-striking-illustration-of

A $200 ChatGPT Pro subscription could represent as much as $14,000 a month in equivalent API-priced usage if pushed to its limits, according to SemiAnalysis — a striking illustration of how heavily flat-rate AI plans can subsidise their biggest users

2+ week, 1+ day ago   (236+ words) The economics of unlimited AI subscriptions are imploding: power users are extracting tens of thousands of dollars in value from plans that cost mere hundreds. By Lab Report · Edited by Mal James Published August 29, 2026 · How we edit The gap grows…...

ScienceBlog.com
scienceblog.com > n-the-hawthorne-effect-is-named-after-a-pattern-in-1920s-factory-lighting-data-that-nobody-checked-against-the-records-for-eighty-years-when-two-economists-finally-found-the-originals

The Hawthorne effect is named after a pattern in 1920s factory lighting data that nobody checked against the records for eighty years — when two economists finally found the originals, the famous result was not in them

2+ week, 2+ day ago   (190+ words) A foundational principle of modern management rests on factory data that was never actually verified—until economists finally checked the archives and found something entirely different. By The Long View Published August 28, 2026 · How we edit That idea, that people alter…...

ScienceBlog.com
scienceblog.com > g-bad-habits-of-intelligent-people

Being smart does not protect you from bias, and can quietly make it worse

2+ week, 3+ day ago   (591+ words) ScienceBlog.com is an independent daily science news publication covering peer-reviewed research across all major scientific disciplines. Founded in 2002, it is one of the longest-running science news sites on the internet. It curates and reports on breakthroughs from universities, research…...

ScienceBlog.com
scienceblog.com > k-the-human-brain-is-so-desperate-to-find-patterns-that-it-will-reliably-see-faces-in-clouds-hear-words-in-static-and-detect-meaning-in-pure-randomness-and-theres-a-name-for-the-glitch-that-makes-it

The human brain is so desperate to find patterns that it will reliably see faces in clouds, hear words in static, and detect meaning in pure randomness, and there's a name for the glitch that makes it happen.

3+ week, 5+ day ago   (259+ words) Our pattern-seeking brains are so hardwired for recognition that they manufacture faces from clouds, static, and toast—a quirk that once kept us alive but now plays tricks on us daily. Published August 19, 2026 · How we edit There’s a whole different…...

ScienceBlog.com
scienceblog.com > t-starcloud-first-llm-trained-orbit-nvidia-h100-500-km

In December 2025, startup Starcloud trained the first large language model ever trained in orbit, using an NVIDIA H100 — the same class of GPU built for Earth's AI data centres, now running roughly 500 kilometres above the planet.

4+ week, 13+ hour ago   (794+ words) Starcloud-1 used a data-centre-class NVIDIA H100 to train NanoGPT on Shakespeare in orbit, a small but genuine demonstration distinct from its preloaded Gemma inference run. By Lab Report · Edited by Lachlan Brown Published August 16, 2026 · How we edit On 10 December 2025, engineers at…...

ScienceBlog.com
scienceblog.com > sb-p-quote-by-viktor-frankl-between-stimulus-and-response-there-is-a-space-in-that-space-is-our-power-to-choose-our-response-and-in-our-response-lies-our-growth-and-our-freedom

Quote by Viktor Frankl: Between stimulus and response there is a space, in that space is our power to choose our response, and in our response lies our growth and our freedom

4+ week, 1+ day ago   (1009+ words) Viktor Frankl's line about the space between stimulus and response is quoted everywhere, but the psychology behind it is more specific than the poster version suggests. We read the research on autonomy, self-determination theory, and what the pause can and…...