Tag: inteligencia artificial

Meta becomes the first customer of Arm’s new custom chips

ARM, una de las empresas más importantes en el campo de los procesadores para smartphones, está considerando una evolución significativa en su modelo de negocio. La compañía, conocida por diseñar la arquitectura y los núcleos de los procesadores que alimentan tanto los smartphones Android como los iPhone, planea expandirse a la fabricación de semiconductores, con el lanzamiento de su primer chip programado para este verano. Este paso marca una nueva era para ARM, especialmente con la noticia de que Meta será el primer cliente en utilizar estos nuevos semiconductores en sus centros de datos. Meta y ARM comienzan […]

ARM, one of the most important companies in the field of processors for smartphones, is considering a significant evolution in its business model.

Facebook DOWNLOAD

The company, known for designing the architecture and cores of the processors that power both Android and iPhone smartphones, plans to expand into semiconductor manufacturing, with the launch of its first chip scheduled for this summer.

This step marks a new era for ARM, especially with the news that Meta will be the first customer to use these new semiconductors in its data centers.

Meta and ARM Begin a New Era in the World of AI

This collaboration not only underscores the growing importance of artificial intelligence, but also puts ARM in a competitive position against giants like Qualcomm, which is currently in a legal dispute with the British firm.

The new chip, which will be aimed at servers and will be customizable, will allow companies like Meta to adapt the hardware to their specific needs. This may lead to tensions in the relationship between ARM and some of its main clients, as the company seems to intend not only to design but also to manufacture and market its own chips.

The production of the ARM chip will be carried out in collaboration with TSMC, a renowned Taiwanese manufacturer, highlighting the importance of partnerships in the supply chain.

In addition, SoftBank’s ambitious $500 billion plan, which includes OpenAI and aims to develop artificial intelligence infrastructure, could pave the way for broader implementations of these new chips in other devices.

Facebook DOWNLOAD

On the horizon, ARM could also venture into creating chips for a niche still in development: AI-powered personal devices, which adds a level of uncertainty and expectation to its upcoming launch.

OpenAI's rival AI has started to arrive in cars: DeepSeek R1 comes from Nissan

Nissan has taken a significant step in the competitive electric vehicle market in China with the launch of its new N7, which becomes the brand’s first electric car to incorporate DeepSeek R1 technology. This model, presented at the Guangzhou Auto Show, aims to revitalize Nissan’s presence in the country, where it faces significant challenges against industry leaders like BYD and XPeng, who are quickly capturing market share. The N7, designed on Dongfeng’s new modular platform, has dimensions that make it slightly larger than the Tesla Model […]

Nissan has taken a significant step in the competitive electric vehicle market in China with the launch of its new N7, which becomes the brand’s first electric car to incorporate the DeepSeek R1 technology.

ChatGPT DOWNLOAD

This model, presented at the Guangzhou Auto Show, aims to revitalize Nissan’s presence in the country, where it faces significant challenges against industry leaders like BYD and XPeng, who are quickly capturing market share.

The N7, designed on Dongfeng’s new modular platform, has dimensions that make it slightly larger than the Tesla Model 3, measuring 4,930 mm in length, 1,895 mm in width, and 1,487 mm in height.

How DeepSeek is integrated with the Nissan car

According to Dongfeng Nissan’s vice president, Zhou Feng, the integration of DeepSeek allows cars to better understand the driver’s intent, which promises a smoother driving experience and a more effective interaction between human and machine.

With a focus on providing a stress-free driving experience and superior comfort, the N7 is also equipped with advanced autonomous driving features, including Momenta’s “Navigate on Autopilot” technology, which offers high-speed navigation and smart parking.

As the N7 seeks to position itself against rivals like Tesla’s Model 3, its success will be crucial for Nissan to regain ground in the Chinese market, especially after reporting a 15% drop in its sales in the country last year.

ChatGPT DOWNLOAD

Despite the hopes generated by the N7, Nissan has not yet revealed prices or an official launch date. Analysts are wondering if this new model will help the brand reverse the negative trend and face the fierce competition from local companies in the electric vehicle sector.

The TikTok team launches two new AIs that generate video and are crazy

Bytedance, the company known for owning TikTok, has recently launched two artificial intelligence models that promise to revolutionize audiovisual content generation. The new models, OmniHuman-1 and Goku, are capable of creating high-quality videos from images and audio clips, achieving impressive lip synchronization and image resolution. They are competitors to Sora from OpenAI. OmniHuman-1 stands out for its multimodal capability, effectively combining video, audio, and animations. The quality of the generated content is remarkable, which can be verified through a series […]

Bytedance, the company known for being the owner of TikTok, has recently launched two artificial intelligence models that promise to revolutionize audiovisual content generation.

ChatGPT DOWNLOAD

The new models, OmniHuman-1 and Goku, are capable of creating high-quality videos from images and audio clips, achieving impressive lip synchronization and image resolution. They are the competition of Sora, from OpenAI.

This is how OmniHuman-1 and Goku are

OmniHuman-1 stands out for its multimodal capability, effectively combining video, audio, and animations. The quality of the generated content is remarkable, which can be verified through a series of demonstration videos on its GitHub page.

In addition, this model is not limited to photorealistic videos; it can also generate cartoons, animated objects, and complex poses, which expands its potential applications in digital art.

On the other hand, the Goku model, with 8 billion parameters, is aimed at the advertising market. Leveraging Bytedance’s vast media library, Goku produces ads targeted at social media platforms, demonstrating that the company has a clear strategy to integrate its AI technology into the advertising sector.

The first produced clips reflect this approach, with examples of cosmetic products and other commercial items.

Despite their impressive capabilities, both models are still in the laboratory phase and require a tremendous amount of computing power, which limits their current availability.

However, the film industry must prepare for an imminent change, as it is believed that AI could displace workers in audiovisual production, starting with the lesser roles.

ChatGPT DOWNLOAD

In summary, Bytedance is positioning itself as a key player in the race to dominate video generation technology through artificial intelligence, which could have a profound and accelerated impact on the entertainment industry.

OpenAI plans to launch its custom AI processor by 2026: goodbye to Nvidia?

OpenAI plans to finalize its first design of a custom artificial intelligence processor in the coming months, sending it to TSMC for production, with the goal of achieving large-scale manufacturing by 2026. This move aims to ensure the company’s competitiveness against giants like Google, Meta, and Microsoft, which have already made progress in developing their own custom silicon. The design of OpenAI’s processor will include a systolic array, a structure that allows for efficient matrix or vector calculations by connecting processing elements in a pipeline configuration. How will the […]

OpenAI plans to finalize its first design of a custom artificial intelligence processor in the coming months, sending it to TSMC for production, with the goal of achieving large-scale manufacturing by 2026.

ChatGPT DOWNLOAD

This movement aims to ensure the company’s competitiveness against giants like Google, Meta, and Microsoft, which have already made progress in the development of their own custom silicon.

The design of OpenAI’s processor will include a systolic array, a structure that allows for efficient matrix or vector calculations by connecting processing elements in a pipeline configuration.

How the ChatGPT chip will compete with Nvidia

This processor is expected to use HBM memory and be manufactured using TSMC’s capacity in the N3 process, which belongs to the 3 nm classification.

OpenAI is collaborating with Broadcom, where the artificial intelligence company will develop the key intellectual property while Broadcom will handle the assembly of the final design.

Richard Ho leads this project, who has experience in developing TPUs for Google. Although the OpenAI team has grown to 40 engineers, it is still considerably smaller compared to the teams at Amazon or Google, suggesting that its internal contribution to the design may be limited.

The development of high-performance processors is a very costly project, with estimates suggesting that each model could cost hundreds of millions of dollars, not including the necessary infrastructure that could double that figure.

However, the growing demand for AI chips has led companies like Meta and Microsoft to allocate multi-billion dollar investments in infrastructure, which could encourage OpenAI to follow a similar path.

ChatGPT DOWNLOAD

If OpenAI manages to achieve mass production of its processor by mid-2026, it could begin its deployment in the second half of that year.

The ChatGPT ad for Super Bowl 2025 was made 100% by humans

OpenAI presented its first TV ad during the Super Bowl of 2025, a 60-second spot promoting ChatGPT, its chatbot and most successful product. The cost of the ad was 14 million dollars. This ad, titled “The Age of Intelligence,” is notable for being created by humans, 100%, without the intervention of any artificial intelligence. The animation, which adopts a pointillism style in black and white, illustrates the greatest inventions of humanity throughout history, starting with a pulsating dot and progressing through innovations such as the […]

OpenAI presented its first TV commercial during the Super Bowl of 2025, a 60-second spot promoting ChatGPT, its chatbot and most successful product. The cost of the ad was 14 million dollars.

ChatGPT DOWNLOAD

This announcement, titled “The Age of Intelligence,” stands out for being created by humans, 100%, without the intervention of any artificial intelligence.

The animation, which adopts a black and white pointillism style, illustrates the greatest inventions of humanity throughout history, starting with a pulsating dot and progressing through innovations such as fire, space exploration, and the arrival of the internet, culminating in the introduction of artificial intelligence.

A different announcement, a 10 for OpenAI

The marketing director of OpenAI, Kate Rouch, indicated that the purpose of the announcement was to highlight practical applications of artificial intelligence, rather than focusing on abstract concepts such as artificial general intelligence (AGI) or “superintelligence”.

Rouch emphasized that the announcement was designed to resonate with a broad audience, many of whom may not be familiar with AI. The commercial is estimated to reach 130 million viewers just in the United States, thus expanding the audience for ChatGPT.

Despite expectations that OpenAI would use its own tools, such as the video generator Sora, for the creation of the advertisement, Rouch explained that these were only used to prototype ideas. “This is a celebration of human creativity,” he stated, justifying the decision not to involve AI in the final production process of the advertisement.

The ChatGPT commercial was one of the highlights of Fox’s broadcast during the Super Bowl, alongside Kendrick Lamar’s halftime show. During this event, the CEO of OpenAI also shared a new post on his blog, in which he highlighted three observations about AGI and its potential to benefit humanity.

Meta's new brain-computer interface has an 80% accuracy, and Neuralink should be nervous

Meta has introduced an innovative device capable of translating brain activity into text, achieving a remarkable accuracy of 80% in identifying letters through neuronal activity. However, this advancement comes with significant limitations: the device, which weighs approximately half a ton and costs 2 million dollars, requires a special environment to operate and the user must remain completely still during its use. The system, developed by Meta’s artificial intelligence and neuroscience teams, analyzes brain activity and, based on the magnetic signals emitted by neurons, can determine which keys […]

Meta has introduced an innovative device capable of translating brain activity into text, achieving a remarkable accuracy of 80% in identifying letters through neuronal activity.

Discord DOWNLOAD

This advancement, however, comes with significant limitations: the device, which weighs approximately half a ton and costs 2 million dollars, requires a special environment to operate and the user must remain completely still during its use.

The system, developed by artificial intelligence and neuroscience teams at Meta, analyzes brain activity and, based on the magnetic signals emitted by neurons, can determine which keys are pressed, thus interpreting a person’s thoughts.

How this precise Meta interface works

A través de un escáner de magnetoencefalografía (MEG), se recopilan hasta 1,000 instantáneas de la actividad cerebral por segundo, lo que permite a un modelo de IA, llamado Brain2Qwerty, identificar las letras correctas que se están “escribiendo” mentalmente.

While the system shows impressive potential for improving the understanding of brain function, its current practical application is limited. Users need to be in a specially protected room to block the Earth’s magnetic field, which is trillions of times stronger than the signals emitted by the human brain. Additionally, any slight movement of the head disrupts the accuracy of the reading.

Despite the device’s restrictions, the non-invasive approach of this research offers a less dangerous alternative to more intrusive technologies, such as those developed by Neuralink, which require brain implants.

La investigación también podría contribuir a tratamientos para lesiones y enfermedades neurológicas, abriendo nuevas oportunidades en el campo de la neurociencia.

Discord DOWNLOAD

Historically, technology tends to become more accessible over time. While today this machinery seems enormously complicated and expensive, it is plausible that, with the advancement of research, what is now laboratory equipment could become more portable devices in the future.

Elon Musk declares war on Sam Altman: presents a purchase offer for OpenAI for 97 billion dollars

In a escalation of the hottest rivalry in Silicon Valley, Elon Musk has just announced that he is leading a $97.4 billion bid to buy the nonprofit branch of OpenAI, according to The Wall Street Journal. According to the American outlet, the bid was delivered to the board on Monday morning. The offer brings together a powerful coalition of backers, including Musk’s own AI company, xAI, and heavyweights from venture capital such as Valor Equity Partners, Hollywood mogul Ari Emanuel, and the venture capital firm 8VC, […]

In an escalation of the hottest rivalry in Silicon Valley, Elon Musk has just announced that he is leading a $97.4 billion bid to buy the nonprofit branch of OpenAI, according to The Wall Street Journal. According to the American outlet, the bid was submitted to the board on Monday morning.

ChatGPT DOWNLOAD

The offer brings together a powerful coalition of sponsors, including Musk’s own AI company, xAI, and heavyweights of venture capital such as Valor Equity Partners, Hollywood mogul Ari Emanuel, and the venture capital firm 8VC, co-founded by Palantir’s Joe Lonsdale.

This offer comes at a crucial time for OpenAI. The CEO, Sam Altman, is already juggling multiple massive deals: turning the company into a for-profit structure, raising $40 billion with a valuation of $340 billion, and launching a $500 billion AI infrastructure project.

From lawsuits to the attempted purchase, this is how Elon Musk has changed the strategy

For Musk, who co-founded OpenAI with Altman in 2015 before leaving in 2019, this offer represents his boldest move to date to challenge what he considers a betrayal of the company’s original mission to develop AI safely and openly (an accusation for which he is currently suing OpenAI).

“It is time for OpenAI to be the open-source and safety-focused force for good that it once was,” Musk said in a statement provided by his attorney Marc Toberoff to The Wall Street Journal. “We will ensure that happens.”

The offer also adds an important layer of complexity to the profit-driven conversion anticipated by OpenAI, as Musk’s team has promised to match or exceed any higher offer. This creates a tricky situation for Altman, who is already negotiating with Microsoft and other stakeholders about the equity in the proposed profit structure.

Shortly after the news was announced, Altman tweeted: “No, thanks, but we will buy Twitter for $9.74 billion if you want.”
OpenAI pointed out Altman’s post when asked for a comment. Toberoff did not immediately respond to a request for comments.

From ethics to war: the shift in Google's approach to AI development

Google has made a significant shift in its artificial intelligence (AI) development policy, removing its previous commitment not to engage in the creation of dangerous technologies, including weapons. This change comes in a context where the company had promised, in earlier versions of its AI principles, that it would not develop technology whose main purpose was to cause harm to people or facilitate surveillance that violated internationally accepted norms. Google is compelled to bet on business, against its foundational values. In 2018, Google refused to renew its contract with the government for ‘Project Maven’, […]

Google has made a significant shift in its policy regarding artificial intelligence (AI) development, removing its previous commitment not to engage in the creation of dangerous technologies, including weapons.

Gemini DOWNLOAD

This change occurs in a context where the company had promised, in previous versions of its AI principles, that it would not develop technology whose main purpose was to cause harm to people or facilitate surveillance that violated internationally accepted standards.

Google forced to bet on business, against its foundational values

In 2018, Google declined to renew its contract with the government for ‘Project Maven’, which focused on drone surveillance analysis, and did not participate in a cloud contract related to the Pentagon, citing ethical concerns.

However, in 2022, it was discovered that Google’s involvement in ‘Project Nimbus’ had raised concerns about the possible facilitation of human rights violations, leading to internal tensions and protests among the company’s employees.

The CEO of Google DeepMind, Demis Hassabis, has stated that democracies must lead the development of AI in an environment where there is fear of an arms race.

His statement comes amid fears that AI technology could be used in warfare, such as in the creation of super soldiers, which resonates with the idea of a possible global confrontation, especially between the United States and China.

As the United States government seeks to partner more with tech giants to enhance its military capabilities, there are questions about whether technology should focus on improving the world or on maximizing profits through government contracts.

Gemini DOWNLOAD

As Google seeks new opportunities in this field, the international community watches with concern the ethical and social implications of these developments.

AI has just achieved another milestone: it can now distinguish whiskies better than you

Another victory for artificial intelligence. An AI-powered algorithm has identified the five strongest notes of each analyzed beverage with more precision than any member of a panel of expert tasters. Researchers have used this technology to predict the notes that whisky releases and to determine whether a drink has been made in the United States or in Scotland, two of the world’s largest whisky-producing regions, with differences ranging from the type of production to the ingredients used. The work represents a step forward towards automated systems capable of predicting the complex aroma […]

Another victory for artificial intelligence. An AI-powered algorithm has identified the five strongest notes of each analyzed beverage with more precision than any member of a panel of expert tasters.

Subscribe to the Softonic newsletter and get the latest in tech, gaming, entertainment and deals right in your inbox.

Subscribe (it's FREE) ►

The researchers have used this technology to predict the notes that whisky releases and determine whether a drink has been made in the United States or Scotland, two of the places where the most whisky is produced in the world, with differences ranging from the type of production to the ingredients used.

The work represents a step forward towards automated systems capable of predicting the complex aroma of whisky from its molecular composition. Expert panels often evaluate woody, smoky, buttery, or caramelized aromas, which can help ensure they do not vary substantially between batches of the same product.

How AI Can Help Master Blenders

Determining the aroma of a whisky is not an easy task. Most of the stronger notes of the spirit are a complex mixture of chemical substances that interact in the nose and even mask each other to create a particular aromatic impression. These interactions make it extremely difficult to predict how the whisky will smell based on its chemical signature.

For the latest work, the researchers obtained the chemical composition of 16 American and Scottish whiskies, including Jack Daniel’s, Maker’s Mark, Laphroaig, and Talisker, and details of their aromas from a panel of 11 experts. The information was used to train artificial intelligence algorithms to predict the five main aromas and the origin of the beverages from their molecular components.

One of the algorithms had an accuracy of over 90% in distinguishing American alcoholic beverages from Scottish ones, although the performance would likely decrease with beverages it had not been trained on.

On average, it identified the five strongest notes of each whisky with greater accuracy and consistency than any of the expert panel members. The results have been published in Communications Chemistry.

The compounds menthol and citronellol helped identify American whiskies, which often have a caramel note. Methyl decanoate and heptanoic acid were important for identifying Scotch whisky, which usually has a smoky or medicinal aroma.

ChatGPT DOWNLOAD

Researchers see applications that go beyond whisky, from detecting counterfeit products through discrepancies in their smell to finding the best way to mix old recycled plastics, which can develop unpleasant odors, into new products without the odor being noticeable.

This AI robot is the Tamagotchi of the future: Introducing Mixi Romi Lacatan

Si tienes algunas canas como yo, seguramente recuerdas los Tamagotchis, esos pequeños robots que venían de Japón y eran una especie de mascotas digitales en miniatura. La razón para recordarlos es que la invención de hoy nos recuerda mucho a ese producto divertido y exótico. Siendo técnico, la invención que… Continuar leyendo “”

If you have a few gray hairs like me, surely you remember the Tamagotchis, those little robots that came from Japan and were a kind of miniature digital pets. The reason for remembering them is that today’s invention reminds us a lot of that fun and exotic product.

Subscribe to the Softonic Newsletter and receive the latest news in tech, games, entertainment in your inbox

Subscribe (it's FREE) ►

Technically speaking, the invention we are discussing is the conversational robot Mixi Romi Lacatan, which has been awarded the CES 2025 Innovation Award.

The latest Mixi model incorporates several AI-based features, such as the ability to see and talk about its surroundings. And we say latest model because, as we will now explain, its debut took place three years ago.

A robot that acts as a pet and a family member – this is Romi

This daily companion helps singles and seniors who don’t have pets, friends, or family to pass the time alone, explain the company responsible for creating this huge AI-powered tamagotchi.

The original Romi was launched in 2021 with the aim of “healing like a pet and understanding like a family member.” Some of the people surveyed during the COVID-19 pandemic lockdowns felt lonely and wanted to have pets, but couldn’t due to housing restrictions imposed on pet ownership.

And that’s where Romi comes in, which satisfies that need by engaging in daily conversations in English or Japanese with its user and even waking up sleepy users with its programmable alarms.

The latest model adds four AI-based functions to Romi. The robot can remember conversations, events, and memories from the past thanks to long-term memory powered by cloud storage. The conversation now flows naturally with the pace set by the human interlocutor, even when interrupted.

Synthesized voices, but similar to human ones, can be adjusted to the owner’s preferences. And the robot can see the world around it and use information about the world in its conversations.

Other AI features include the integration of OpenAI GPT-4o in assistance mode, which allows Romi to provide detailed responses, including cooking recipes or detailed scientific answers. This mode also allows the robot to be set in one of three scenes (business, teacher, and simulation game) to offer more appropriate responses.

ChatGPT DOWNLOAD

Romi Lacatan costs 89,800 yen (580 dollars) and the subscription to artificial intelligence services is 1,780 yen per month (12 dollars per month). Mixi accepts pre-orders in the Romi store, with delivery scheduled for spring 2025 in Japan.