→ Get unlimited parallel test runs
→ Zero Flakes guaranteed
Full-text search across article titles and summaries. Refine with sector, year, content type, and reading-time filters.
→ Zero Flakes guaranteed
Amazon will release a revamped version of its Alexa voice assistant in October. It will be powered by AI models from Anthropic's Claude rather than Amazon's own in-house AI technology, as Amazon's internal software reportedly struggled with performance issues during initial testing. The new Alexa will enable users to engage in more complex, context-aware conversations with the assistant. Amazon plans to offer the upgraded Alexa as a paid subscription service, separate from existing Prime memberships, with pricing likely to be between $5 to $10 per month.
Nvidia's Blackwell chips are about twice as big as its predecessors, housing 2.6 times the number of transistors. Instead of one big piece of silicon, Blackwell consists of two advanced processors and numerous memory components joined in a single, delicate mesh of silicon, metal, and plastic. The manufacturing of each chip has to be close to perfect, presenting engineering challenges that have a sizable impact on the bottom line, with each defect rendering a $40,000 chip useless. This article looks at some of the challenges Nvidia had to overcome to produce the chip.
Semaglutide could be used to slow down the process of aging. A study tracking people aged 45 or older who were overweight and had cardiovascular disease but not diabetes found that those who took the drug died at a lower rate from all causes. It also improved heart failure symptoms and cut levels of inflammation regardless of whether or not people lost weight. The study reinforces the theory that being overweight increases the risk of death due to many etiologies.
Sponsored
Between malicious links and invasive tracking, the modern internet is not safe for unprotected browsing. Surfshark One is everything you need to stay secure in your online adventures: VPN [11], data breach Alert, Antivirus [12], ad blocker, alternative identity, and more.
Valkey is a high-performance key/value datastore forked from Redis. It has gained a lot of support, with two-thirds of former top Redis maintainers and developers switching over to the project. Valkey 8.0 introduces serious speed improvements and new futures that Redis users have wanted for some time, including improved debugging tools, enhanced documentation, and new command-line utilities for easier management and monitoring. It maintains full API and CLI compatibility with Redis and compatibility with the Redis Serialization Protocol, making migrating from Redis to Valkey easy.
Making a jump similar to the 100x jump going from GPT-3 to GPT-4 is going to be very hard. It will require a lot more training data and larger data centers. It is difficult to see whether efforts to improve models will bear any fruit. Policies like SB 1047 threaten to further slow down progress.
Today's AI may seem harmless, but these systems are being woven into society in ways that may be hard to untangle. It could be problematic for people if AI has somehow trained a bias against them, for example, a person whose name was associated with a negative event may be unfairly discriminated against by other chatbots. Chatbots are highly suggestible - if people add text on the internet and it gets memorized, it's memorialized in the language model.
Physicists like to write books with grand unified theories of the world, but that view of the world is limited and we shouldn't expect a theory or algorithm that explains it all. The universe is made up of many different worlds, both physical and metaphorical, and many of these worlds aren't connected to each other in any meaningful way. The truth is contextual, so it isn't a good idea to reduce everything to fundamentals. Reality being a disjoint collection of domains seems infinitely more rich than it being a single equation.
Sponsored
Online. No set class times. 200+ degrees. Low online tuition. Find your degree at Southern New Hampshire University today. Inquire here
New competition is coming from startups that promise to undercut OpenAI's services with cheaper bots that are also better at certain narrow tasks.
Executives quitting Apple don't actually leave - they are usually moved to other roles, like Luca Maestri, who recently departed his role as Chief Financial Officer but will still oversee the Information Systems and Technology unit.
Scaling a startup doesn't always mean switching to manager mode.
If you don't regularly have people thinking you're stupid, you're either likely having extremely filtered interactions with people, you are avoiding looking stupid, or you are not noticing when people think you're stupid.
This article explains how the creator of One Million Checkboxes discovered a group of teens sending him secret messages using binary.
There are still technical limitations, unknown unknowns, and other issues making establishing causality between the microbiome and disease difficult.
Sponsored
Companies spend $4,000+ and take 36 days to fill a position, only to lose 17% of new hires in their first month. In many cases, the problem is onboarding “sink or swim” processes.
Apple and Nvidia are in talks to invest in OpenAI as part of a fundraising round that would value OpenAI at above $100 billion. It is an unusual move for Apple, as the company doesn't usually invest in startups. Nvidia has stepped up its investment activity in the past two years, putting its money into AI-related companies. OpenAI is one of the largest users of Nvidia's AI chips.
A report from China claiming to use information from Tesla's supply chain states that Tesla is planning a Model Y refresh in the first quarter of 2025. Tesla is also reportedly planning to start production of a 7-seat Model Y at Gigafactory Shanghai soon. It is expected to be an upgraded version of the US seven-seater configuration for the Asian and European markets. Most of the changes made to the Model 3 in last year's refresh will make it to the new Model Y.
The Chinese government's current five-year plan makes comprehensive research facilities for crucial fusion projects a major priority for the country's national science and technology infrastructure. It is estimated to be spending around $1.5 billion each year on fusion, almost double what the US government allocated this year for this research. The country has built itself from being a non-player 25 years ago to having world-class capabilities. Chinese scientists are aiming to build a test reactor with the goal of producing up to 1 gigawatt of fusion power in the 2030s.
Decision making in biotech is ridiculously difficult. ML systems can automate low- and medium-level decision making. These systems can generate more data and better synthesize that data into decisions. They are able to systematize human rational frameworks into quantitative, interpretable, and scalable analytic engines. Enabling model-driven decision making would require building an organization in a data centric way with an engineering culture.
Sponsored
Are slow test cycles bottlenecking your dev teams' release velocity? With QA Wolf [10], your organization can run entire test suites in minutes to ship features faster, and with confidence.
Elasticsearch and Kibana are now open source again. AGPL will be added as another license option next to Elv2 and SSPL in the coming weeks. Elastic changed its license 3 years ago to remove a lot of market confusion, but the landscape has changed so it is able to become open source again.
This developer recently crossed over $1 million on GitHub Sponsors after leaving their day job 5 years ago with no plan. In this article, they provide the strategy they followed to reach that achievement. The developer sold screencasts of them building stuff with their product. They also documented their work in detail through writing, speaking, and recordings.
SB-1047 passed the California State Assembly by a 45-11 vote. It now faces just one more procedural state Senate vote before heading to the governor's desk. The bill asks AI model creators to implement a 'kill switch' that can be activated if a model starts introducing novel threats to public safety and security. It has been criticized for focusing on risks from an imagined future AI rather than real present-day harms like deepfakes or misinformation.
The ROI on AI has been positive so far. Scaling laws may eventually collide with economics - if they hold. If scaling laws hold and there are no fundamental technology breakthroughs, then the models trained in 2026/2027 are likely to require clusters that cost well over $100 billion. At that point, the ROI on AI may begin declining.
OpenAI and Anthropic have signed deals granting the US government early access to conduct safety testing on their new AI models before they're released to the public.
Departure mono is a monospaced pixel font inspired by the constraints of early command-line and graphical user interfaces.
Meta's Llama large language models have been downloaded almost 350 million times, with approximately 20 million of those downloads taking place in the last month alone.
This post describes the process of discovering a bug that allowed anyone with a basic knowledge of SQL injection to add anyone they wanted to airport security systems.
ChatGPT now has more than 200 million weekly active users.
Chinese tech professionals are moving to Silicon Valley for opportunities they don't believe are available in China anymore.
Sponsored
☎️ Are phone-based jobs about to go extinct?
NFTs haven't recovered despite crypto values rising. While they're hanging on, the technology is in troubled waters. OpenSea, once the largest marketplace for NFTs, is facing pending litigation from the SEC, a previous unreported 'matter' with the FTC, inbounds from US and international tax authorities, heightened competition, accusations of gender discrimination, and employee attrition. This article tells the story of the marketplace's rise and fall.
OpenAI is reportedly in talks to raise funds at a valuation of more than $100 billion, substantially higher than the company's previous $86 billion valuation. Investors include Thrive Capital and Microsoft. The financing will be OpenAI's biggest outside infusion of capital since Microsoft invested nearly $10 billion in January last year. While OpenAI's revenue eclipsed $3.4 billion early this year, the company is on track to lose nearly $5 billion by year end. It has already burned through $8.5 billion on AI training and staffing.
Workers have been making last-minute upgrades to SpaceX's Starship launch pad in South Texas pretty much every day for the last couple of weeks. If all goes well on the Starship's next test flight, the launch pad's mechanical arms will close together to capture the Super Heavy booster as it descends back to Earth. SpaceX announced that Starship and Super Heavy were ready to fly, pending regulatory approval, on August 8. The FAA is still evaluating SpaceX's proposed flight profile. It is unclear how long approval will take, but the state of the launch pad suggests that the next test is still at least a couple of weeks away.
The competition for AI inference is heating up, particularly in terms of power efficiency. ML Commons has released the results of its latest AI inferencing competition. The round included first-time submissions from teams using AMD Instinct accelerators, the latest Google Trillium accelerators, chips from Untether AI, and Nvidia's new Blackwell chip. This article looks at the results - while Nvidia's chips are still ahead in many categories, chips from other manufacturers are starting to catch up.
Sponsored
In this LinkedIn post, Dan shares 3 key insights he's gained about messaging a developer-focused audience from building ScraperAPI and TLDR. Dan has been actively posting on LinkedIn - follow him for developer marketing content [11].
Using ChatGPT to code can be a real productivity boost. Instead of reading examples on Stack Overflow and adapting them to their particular case, developers can instantly get code tailored to their specific needs. It can make developers more efficient, but they need to remember that it is still a tool. It is important to always test the code it provides to quickly discover any hallucinations.
State is time and time is state. The only way to see the passage of time is by measuring changes in observable state. Changes in state matter in that they create new times. If a state update is purely internal and cannot affect the observable state, then it doesn't advance time. This model can be used to reason about abstract systems.
Nvidia recently reported that its sales and profits had more than doubled during the last quarter. The company also projected that sales in the current quarter will increase by 80% compared to a year ago, exceeding earlier estimates. Nvidia's revenue was $30.04 billion last quarter, and its net income rose to $16.95 billion. It will spend $50 billion on repurchasing its own shares. Nvidia now supplies more than 90% of the chips essential to building AI, but competition is on the rise, with other companies developing custom chips to lessen their dependence on Nvidia.
Telegram CEO Pavel Durov has been indicted in France on charges including web-mastering an online platform in order to enable an illegal transaction in an organized group, complicity in drug trafficking, distribution of child pornography, and providing cryptology services without making required declarations to government officials. Telegram has released a statement saying that it follows the law and industry standards on moderation and that it was absurd to claim that a platform or its owner is responsible for the abuse of that platform. Durov was ordered to post bail of 5 million euros. He is forbidden from leaving the country and must report to police twice a week while the case continues.
Sponsored
Streamline and secure identity, access, devices, and hardware inventory all from one platform. See why 6,000+ IT and security teams