Depois de encerrar o desenvolvimento do Arcno início do ano passado, a The Browser Company anunciou hoje que está encerrando de vez o funcionamento do seu outrora popular navegador, forçando usuários a migrar para alternativas como o Dia — outro browser da casa.
A empresa explicou no site oficial do Arc que o fim do suporte ao navegador acontecerá em fases, culminando na sua morte definitiva em 22 de dezembro (tanto no macOS quanto no Windows).
We've decided to sunset Arc on December 22, 2026.
That's not an easy sentence to write. We spent years building Arc and celebrating each and every release along the way: we have so much pride and love for what we made alongside the Arc community.
A partir de hoje, já não é mais possível criar novas contas no Arc; em 23 de novembro, o recurso Browse for Me e outras ferramentas de inteligência artificial do navegador serão desativadas, bem como a habilidade de “logar e sincronizar as suas abas”. Em 22/12, a The Browser Company também removerá o Arc Search do Google Play e desativará todas as contas de usuários.
Como comentamos no ano passado, a desenvolvedora abandonou o Arc para focar no Dia, que conta com um grande foco em IA. Na época, o diretor executivo da The Browser Company, Josh Miller, disse que sua equipe até considerou tornar o Arc em um navegador de código aberto, mas que desistiu da ideia pois o kit de desenvolvimento de software (SDK, na sigla em inglês) do Arc ainda era valioso para a companhia.
Desde então, o Arc chegou a receber atualizações de segurança — algo que também será encerrado agora. Antes independente, a The Browser Company agora também é uma subsidiária da Atlassian (dona do Trello).
OpenBSD 8.0 is here, marking the project's 61st release and continuing its schedule of semiannual releases dating back to 1996. What started more than 30 years ago as a fork of NetBSD has grown into one of the most security-conscious operating systems in active development. In a post to the project's announcements mailing list, OpenBSD founder Theo de Raadt highlights a host of improvements included with the latest version and writes that the project remains "proud of OpenBSD's record of more than thirty years with only two remote holes in the default install."
The latest OpenBSD version brings new features and improvements across the board. These changes, particularly the new Intel Arc support and expanded virtual address space, should improve compatibility with modern graphics cards and allow the system to map more than 4GB of VRAM. Some of the major updates noted in the release newsletter are as follows:
Added support for Intel Arc A-Series (DG2/Alchemist) graphics cards.
512GB kernel virtual address space on AMD64, up from 4GB, allowing VRAM mapping above 4GB.
The new OpenBSD also retires support for Loongson architecture after the port stopped working with the Clang 22 compiler. If you're interested in reading over all the changes that come with the new OpenBSD 8.0, you can take a look at the release announcement.
OpenBSD 8.0 is here, marking the project's 61st release and continuing its schedule of semiannual releases dating back to 1996. What started more than 30 years ago as a fork of NetBSD has grown into one of the most security-conscious operating systems in active development. In a post to the project's announcements mailing list, OpenBSD founder Theo de Raadt highlights a host of improvements included with the latest version and writes that the project remains "proud of OpenBSD's record of more than thirty years with only two remote holes in the default install."
The latest OpenBSD version brings new features and improvements across the board. These changes, particularly the new Intel Arc support and expanded virtual address space, should improve compatibility with modern graphics cards and allow the system to map more than 4GB of VRAM. Some of the major updates noted in the release newsletter are as follows:
Added support for Intel Arc A-Series (DG2/Alchemist) graphics cards.
512GB kernel virtual address space on AMD64, up from 4GB, allowing VRAM mapping above 4GB.
The new OpenBSD also retires support for Loongson architecture after the port stopped working with the Clang 22 compiler. If you're interested in reading over all the changes that come with the new OpenBSD 8.0, you can take a look at the release announcement.
Researchers from the private nuclear fusion energy company Commonwealth Fusion Systems have published a series of papers that they claim confirm that their flagship ARC reactor will produce more usable energy than it consumes if built as designed.
Although the company projects it will deliver the first commercial fusion plants by the early 2030s, other researchers say there are remaining challenges to overcome, and won’t declare the ARC reactor successful until an operational version is constructed to validate the research team’s theoretical predictions.
ARC Fusion Reactor ‘Will Work’ When Built
Scientists have long dreamed of capturing the nuclear fusion reaction that fuels our Sun and using it to generate clean, renewable power. Private fusion firms like CFS have received nearly $10 billion in funding over the last decade alone.
The first laboratory “net-energy” controlled fusion reaction, which fusion scientists call “ignition,” didn’t occur until 2022, when physicists at the US National Ignition Facility in Livermore, California, successfully produced more energy than was used to generate the reaction. Although brief, that moment has fueled speculation that practical fusion energy generation may be right around the corner.
Now, the Commonwealth team, which includes 58 researchers from the company and partner institutes, has published scientific papers outlining the physics of their ARC reactor, including new models that they say account for the “expected behavior” of the superhot plasma at the heart of the design.
Brandon Sorbom, the company’s chief science officer, said that the published works “confirm that when we build the ARC fusion power plant, it will work”.
Company Claims ARC Will Produce 400 Megawatts of Net Electricity
After nearly a decade of research, CFS, founded in 2018, is currently preparing its prototype for a 2027 demonstration it calls ‘SPARC.’ The company said they expect that trial run, which will demonstrate their technology’s ability to generate more energy than it needs to heat the plasma, will help them fine-tune ARC’s final design.
A peek inside the Commonwealth Fusion Systems magnet production facility. Image Credit: CFS.
According to the Commonwealth-led team, ARC will generate more energy than is required to operate the entire facility, including plasma heating, resulting in a true net-positive power-generation facility. This includes producing approximately 400 megawatts of net electricity, which the team noted is enough to power 280,000 average US homes. For comparison, recent census data indicates that Boston, Massachusetts, near the company’s headquarters in Cambridge, has around 271,000 U.S. homes.
Like SPARC, the ARC reactor uses high-temperature superconducting magnets to compress the plasma, which remains suspended by the magnetic fields, preventing it from damaging the chamber walls and terminating any potential fusion reaction. Under this extreme pressure, the nuclei of heavy hydrogen trapped inside the chamber fuse to form helium, the same reaction that takes place inside the Sun.
This fusion reaction results in the release of energy, primarily high-speed neutrons. At its hottest, the donut-shaped ring of plasma reaches seven times the temperature of the Sun.
Critics Applaud Publication While Highlighting Unproven Concepts and Systems
Tony Roulstone, a nuclear engineer at the University of Cambridge, UK, conceded that the CFS team and its collaborating authors are “some of the best in the fusion business” and noted the company’s ARC reactor concept is the result of “good work over many years.” However, Roulstone also warned of the pressure on researchers in privately funded companies to “claim things before the evidence is fully in place.”
David Hammer, a nuclear engineer at Cornell University in Ithaca, New York, said it is a good sign that the team is publishing at a high volume, but also warned that published papers will not guarantee that the built device will work as intended.
Their modeling “has to be proven correct by a fusion reactor that works as designed”, he adds. If the firm can get SPARC operating successfully next year, “it will be an important step forward”, he says.
Hamner also noted that several of the design’s subsystems require further development before it can be declared complete. This includes addressing how plasma will behave in the real world when it becomes unstable versus its modeled behavior, proving out the process for extracting usable energy from the generated heat, and demonstrating that the reactor is robust enough to operate for long periods.
The scientists also noted the challenges in sourcing tritium, the isotopic fuel the reactor requires to operate. Although CFS has outlined its tritium generation process, Hamner noted that the series of published papers neither details the process nor demonstrates its viability.
“There are candidate blanket materials and blanket designs, but there are many engineering person-years needed” to bring a functional ARC-style plant to fruition, he explained.
“We certainly have to build more on top of that foundation laid out in these papers,” a company spokesperson said, adding that they will “be able to produce tritium for operations and for future power plants.”
Editor’s Note: This article was updated to correct a previous version that described neutrinos, rather than fast-neutrons, in the context of the reporting, and to clarify that the first achievement of a “net-energy” controlled fusion reaction (ignition) occurred in 2022.
Christopher Plain is a Science Fiction and Fantasy novelist and Head Science Writer at The Debrief. Follow and connect with him on X, learn about his books at plainfiction.com, or email him directly at christopher@thedebrief.org.
Most of the Steam Deck imitators on the market right now use AMD silicon, specifically the Ryzen Z-series chips. These are the same chips AMD makes for regular laptops, but with different power settings better suited to a compact handheld system. There are handhelds based on Intel silicon (MSI’s Claw is the main one), but Intel hasn’t yet tried making silicon marketed specifically for that purpose.
Today, the company is throwing its hat in the ring with two Intel Arc G-series processors, which will allow gaming handhelds to leverage the company's genuinely quite good Arc B-series integrated GPUs. Intel says that several Arc G-series handhelds will arrive "starting in June 2026, with broader availability throughout the year." These systems will include a new MSI Claw model, a Predator Atlas 8 from Acer, and a device from OneXPlayer.
Intel normally uses its "Arc" branding for integrated and dedicated GPUs, but in this case, the "Arc" brand encompasses the entire chip, including the CPU, GPU, NPU, and other components.
Login
