Full-text search across article titles and summaries. Refine with sector, year, content type, and reading-time filters.
Meta has confirmed that it is working on a 50,000 kilometers-long, $10 billion+ subsea cable project to connect up the globe in the world's longest subsea cable project.
Both Apple and Google restored TikTok to their app stores in the US on Thursday. Other ByteDance apps were also restored. President Trump has said that he would like the US to own a 50% share in TikTok through a joint venture with other tech companies. Competing social networks have been trying to bank on TikTok's uncertain future by releasing rival features and apps.
Elon Musk's offer to buy OpenAI for $97.4 billion has a clear deadline: May 10. There is a stipulation that the buyers are able to examine OpenAI's financial and business records and interview OpenAI staff before the all-cash transaction is finalized. The offer undermines Musk's legal claims that OpenAI's startup's assets can't be transferred away for private gain. Musk will drop the bid to acquire OpenAI if the board commits to keeping it as a nonprofit.
The UK's Advanced Research and Intervention Agency (Aria) is funding a trial to test a brain-computer interface that directly alters brain activity using ultrasound to boost mood. The technology could revolutionize the treatment of conditions such as depression, addiction, OCD, and epilepsy by rebalancing disrupted patterns of brain activity. The device was developed by a US-based non-profit called Forest Neurotech. The trial will recruit patients who have had a part of their skull temporarily removed so that the device can be tested without having to perform surgery. It will run for three and a half years starting in March, and if successful, the device will move into full clinical trials.
Uber is expanding its partnership with Avride to include autonomous deliveries in Jersey City. Customers in the city can now have their orders delivered by one of Avride's autonomous sidewalk delivery robots. Avride's first robotaxi deployment in the US will be in Dallas this year. The startup has also partnered with Grubhub to bring delivery bots to university campuses across the US.
One of the ultimate aims of protein design is to build entirely new enzymes from scratch. Scientists have traditionally discovered enzymes in nature and then adapted them through trial and error. New advances in AI may eventually lead to ways to create bespoke enzymes from the ground up, but doing so remains a formidable challenge. Many of the protein structural databases used to train models consist overwhelmingly of structures gleaned from 'frozen' images, preventing them from making accurate recreations of dynamic structures.
It's possible to use macOS folder actions to automate various tasks, including converting MOV to MP4 and images to WEBP, downloading videos from YouTube, and more. This article explains how to set these automations up. macOS folder actions turn folders into an interface for any CLI app, making the desktop even more useful than ever.
Arm has signed Meta as a customer and will start making chips this year. The first in-house Arm chip will be unveiled as early as this summer. The semiconductor company usually licenses its chip blueprints, so making its own chips is a notable change in strategy. This will turn some of Arm's existing customers into competitors.
People seem to be confused about what can be concluded about scaling large language models. At this point, it is not plausible to precisely predict how far we are from unlocking all of the remaining functions that large language models are capable of. We don't even know all of the component functions of human intelligence - every time we solve something previously out of reach, it turns out that human-level generality is even further out of reach. The invention of human-level artificial general intelligence will disrupt history, but that point is still pretty far off.
The integration of AI has stifled the adoption of new and potentially superior technologies due to training data cutoffs and system prompt influence.
The large-scale engagement of corporations in open source needs to be treated seriously because it has and will continue to have negative consequences even if every individual involved is participating in good faith.
Meta CTO Andrew Bosworth said employees were free to quit if they think that it's okay to leak to the media because of policy disagreements.
Apple may soon allow third-party retailers to sell its Vision Pro headset.
It may be possible to have 'a country of geniuses' within a data center by as early as 2026 - companies will need to be prepared for this future.
This post tells the story of how a doomed open-source project was turned into a fast-growing AI company.
Sponsored
Spencer Rascoff co-founded Zillow. Austin Allison sold his company to Zillow for $125M. Now they've teamed up with investors like SoftBank and Maveron to transform the $1.3T vacation home market – and you can join them for just $2.70 a share [4].
Sam Altman published OpenAI's GPT-4.5 and GPT-5 roadmap on X on Wednesday. Altman acknowledges that OpenAI's product lineup has gotten complicated and that the company wants to simplify its offerings. GPT-4.5, which was internally called Orion, will be OpenAI's last non-chain-of-thought model. GPT-5 will integrate a lot of OpenAI's technology, including o3. Free ChatGPT users will get unlimited GPT-5 chat access at the standard intelligence setting - Plus and Pro subscribers will have access to higher levels of intelligence. It is unknown when the models will be released.
Apple is reportedly exploring both humanoid and non-humanoid robotic form factors. The company is still in an early proof-of-concept stage. 2028 would be an optimistic timeline for mass production. The robot would be part of a future smart home ecosystem.
Google DeepMind has published a video demonstration of miniature humanoid robots playing a one-on-one game of soccer. The robots are able to understand the rules of the game and can anticipate where the ball is going or when to block shots from an opponent. The 15-second video is available in the article.
A team building a detector on the Mediterranean Sea got extremely lucky recently when it picked up the most energetic neutrino ever detected. The detector was only 10% of the size it will be on completion. The neutrino originated outside of our galaxy. Once completed, the detector will be able to help scientists detect where sources of neutrinos reside.
These days, a cheap NVME SSD can supply data much faster than a query interpreter can process it. Depending on the query, removing the interpreter bottleneck can yield several orders of magnitude speedup. The article looks at different options for compiling queries and their pros and cons. Queries might only be run once, so it's only worth spending time on optimization at compile-time if doing so will save even more time at runtime.
This article explores the benefits and trade-offs of building web pages with Tailwind. Tailwind is a CSS framework that uses a utility-first approach that involves styling elements directly without needing to leave HTML. It promises developers the ability to build modern web pages quickly and simply, but there can be significant impacts on maintenance and scaling. Tailwind's utility-first approach can work well for small-to-medium projects where speed and cost-efficiency are top priorities, but for larger projects or teams with established CSS practices, it introduces challenges around readability, maintainability, and scalability.
Working at a large technology company has been an excellent financial decision for technically capable people over the last few years. Joining a rocketship startup - at the right time - can also be an excellent decision. Early-career cash pay at YC startups is often similar to big tech firms. Founding engineers can receive equity that, vested over four years, could potentially be not too different from an offer from Facebook. The median startup fails, making the equity worthless, but there is the potential for extraordinary returns. Being a founding engineer essentially means taking a pay cut and accepting illiquid equity in exchange for a lottery ticket - it's a high-stakes game of calculated risk-taking where due diligence can help tilt the odds in your favor.
Micro LED, one of the most anticipated display technologies, is still years away from being suitable for the mass production of consumer products. The industry is struggling to manage obstacles Luke manufacturing costs and competition from other advanced display technologies. It will probably still be another five years before consumer products are available.
Sponsored
When companies grow, they find themselves spending much more time and resources on communication, but somehow communicating less. Learn how asynchronous video communication can fix the communication breakdown.
It is possible to encode data in any Unicode character.
Anduril Industries will take over Microsoft's augmented reality headset program with the US Army, pending approval from the Department of Defense.
Netflix Bites has opened at the MGM Grand Hotel & Casino - it will operate for a year, offering breakfast, lunch, and dinner options inspired by Netflix series.
It was possible to leak a Google account identifier by blocking them on YouTube - the exploit has been fixed.
FullSoak is a no-build TypeScript full-stack framework for building fast web applications.
It is possible that data speeds are approaching 'fast enough' for most people for most purposes.
Sponsored
It's never been easier to experience the power of AI-assisted coding. Simply open Copilot in your IDE [4], choose your AI model, and you're good to go.
Elon Musk's offer to buy OpenAI for $97.4 billion, backed by a consortium of investors, could force OpenAI's board of directors to reassess how it is valuing the nonprofit. OpenAI is currently negotiating how much equity Microsoft should get in its for-profit company - satisfying all those parties was already complicated. The company must get approval for its conversion from attorneys general in California and Delaware - these officials will oversee the valuation of the nonprofit's assets during a sale to ensure it is properly compensated. Sam Altman has stated that he thinks Musk's offer is just a tactic to slow OpenAI down.
Google appears to be ditching the camera bump this year for the Pixel 9A - it's still there, but just barely. A new set of renders has leaked showing the phone in black, white, pink, and purple. The back cameras will include a 48-megapixel main shooter and a 13-megapixel ultrawide. The phone will have a 13-megapixel selfie camera in the front and a 5,100mAh battery. The base model of the phone, which will have 128GB storage, will cost $499. The Pixel 9A will be released on March 26.
There are now three people with Neuralink implants. Combined, these patients have had their Links implanted for over 670 days and used Telepathy for over 4,900 hours. All three individuals are unable to move their arms and legs. This post looks at how their Links have improved their lives. It tells each of their stories and shows examples of what their Links have enabled them to do.
Scientists in Argentina have created genetically edited horses with increased explosive speed. The five foals have been tweaked to one day be able to outrun the champion mare they received genes from. Gene editing is banned by the International Federation for Equestrian Sports and the British Horseracing Authority, but the biotech firm that created the foals argues that the new horses' genes could have theoretically occurred naturally through selective breeding or genetic mutations.
Sponsored
If QA is a bottleneck on your software engineering team and you're releasing slowly because of it, you need to check out QA Wolf [10].
A range of common patterns is emerging as software products that use generative AI are moving from proof-of-concept to production systems. This article looks at some of the regular patterns used to deal with issues in moving to production. It covers direct prompting, embeddings, evals, hybrid receivers, query rewriting, and retrieval augmented generation. The article provides an example and instructions for when to use the pattern for each pattern described.
This article provides a high-level overview of Meta's hyperscale infrastructure. Meta's engineering culture values moving fast, technology openness, research in production, and shared infrastructure. It has adopted continuous deployment universally and enabled developers to write serverless functions rather than traditional service code. Meta uses hardware-software co-design at the data center scale and autonomously optimizes resource allocations to reduce hardware costs. Its strategy involves co-designing the entire stack.
This year's Exceptional Startups List contains 240 startups, up from 140 last year. 43% of the startups came from the Bay Area. Over 30 of the top venture firms submitted names this year, including some of the best funds on Earth. AI was the main driver of the size of the list. A link to the list is available in the article.
TV screens have overtaken mobile as the primary device for viewing YouTube in the US. Many of the world's biggest events, including elections, the Olympics, Coachella, the Super Bowl, and the Cricket World Cup, play out on the platform. YouTube plans to roll out more tools to support podcasters, improve monetization for creators, and make it easier to discover podcasts. It is making auto-dubbing available for all creators in the YouTube Partner Program later this month. The future will expand to more languages throughout the year.
Sponsored
Watch QA Wolf AI [18] generate a Playwright test for Figma's prototyper in just 6 minutes compared to 29 minutes without AI.
Hopes that return-to-office mandates will bail out the office sector seem premature.
China's Long March 8A is designed to deploy large numbers of satellites for China's massive space-based internet networks.
DeepMind has published research that discusses how to distribute the training of models with billions of parameters among clusters of computers (that could theoretically be widely separated) while producing the same level of quality as before.
WebAssembly is a true write-once-run-anywhere experience, and you can compile several languages into WebAssembly already.
This post walks through how to build a setup that can run medium-sized models completely under your control.