Intel working on ChatGPT-like apps for users

0

The AI boom is reportedly catching up with almost every tech company, especially rival chipmakers like Nvidia and Broadcom. Chipmaker Intel is reportedly working with multiple consulting firms to build ChatGPT-like apps for customers.

The Information reported that Intel is also selling its specialized AI, app-building software directly to corporate clients, citing sources.

According to the report, “Intel began the project with Boston Consulting Group (BCG) earlier this year but has added further consulting-firm partners.”

Intel has not yet responded to the report.

BCG and Intel announced a strategic partnership in May of this year to enable generative artificial intelligence (GenAI) using end-to-end Intel AI hardware and software. This partnership will enable enterprise clients to access fully custom and proprietary solutions while maintaining the security of their sensitive data in isolated, trusted environments.

Employees at BCG were able to retrieve and summarize information via semantic search that had previously been buried in lengthy lists of multi-page documents thanks to a custom natural language chatbot interface powered by Intel AI hardware and software.

“In order for enterprises to safely benefit from the technology, generative AI requires a truly democratized approach that enables more secure and scalable choice,” stated Sandra Rivera, executive vice president and general manager of Intel’s Data Center and AI Group.

In a previous statement, Rivera stated, “Our partnership with BCG enables us to assist clients in developing generative AI applications that necessitate technology optimized throughout the whole stack entirely inside their selected security perimeter.”

In the midst of an AI push, Intel is purportedly getting ready for Windows 12 with its Meteor Lake desktop platform, hinting at a “Windows refresh” for 2024.

When the 5th generation of Intel Xeon processors are released on December 14, they will offer the world’s data centers a combination of faster memory and performance enhancements while consuming the same amount of power, according to a recent company preview.

As stated by Intel, artificial intelligence (AI) is bringing about the “Siliconomy,” a new era of worldwide expansion propelled by the magic of silicon and software in which AI computers will rule our lives.

Audio and Video calls now in X

0

The Elon Musk-owned X Corp has begun to roll out support for audio and video calls in an attempt to compete with Meta-owned WhatsApp.

When opening the social media app, a number of X (formerly Twitter) users saw a notification saying, “Audio and video calls are here!”

In the app’s settings, there’s also a new toggle labeled “Enable audio and video calling.” According to the statement, you can “turn the feature on and then select who you’re comfortable using it with,” The Verge reported.

“Early version of video and audio calling on X,” wrote Musk in a post.

Before the new feature launched, X posted an enigmatic message to its users asking them to get ready for something.

Calls from verified users, individuals you follow, people in your address book, or any combination of the three are now possible with the new feature.

Video calls will be available on the platform as part of X CEO Linda Yaccarino’s announcement last month that the platform will become a “everything app.” “You’ll be able to make video chat calls without having to give your phone number to anyone on the platform,” stated the CEO of X, in a forthcoming statement.

Musk always wanted for Twitter to evolve into “an everything app,” akin to WeChat in China. Two new paid premium tiers from X will launch soon, one of which will have advertisements and cost less than the $8 monthly fee it currently charges. According to the billionaire, there will be an additional, more costly tier that eliminates all advertisements.

“X Premium plans will soon include two additional tiers. The owner of X stated that one costs more but has no ads, while the other costs less and has all the features but no fewer ads.

Google takes action on disruptive notifications and “offensive” AI apps

0

Google has made it clear to developers that all applications—including those created by AI content generators—must abide by its current developer policies, which forbid the creation of content that supports “deceptive behavior” and other types of restricted content, such as child sexual abuse material (CSAM).

Updates to the developer policies were announced by the company in an effort to improve the caliber of apps available on Google Play.

Google stated that it wants to assist in making sure AI-generated content is safe for people and that their feedback is taken into consideration, in keeping with its commitment to responsible AI practices.

The internet behemoth released a statement saying, “We’ll be requiring developers to provide the ability to report or flag offensive AI-generated content without needing to exit the app” early next year.

As a reminder, Google stated that apps that use AI to create content must also abide by all other developer policies.

Certain app permissions have extra restrictions and need to be reviewed by the Google Play team in order to protect user privacy.

“We’re expanding these requirements, including a new policy to reduce the types of apps allowed to request broad photo and video permissions,” the company said. “We’ve found this has been an effective strategy to protect people’s privacy.”

Apps will only be allowed to access images and videos for reasons directly relevant to their operation under the new policy.

Applications are asked to use a system picker, like the Android photo picker, if they only occasionally or once in a while need to access these files.

Stronger restrictions are also placed on the use of full screen intent notifications, which convey urgent messages and demand the user’s immediate attention.

Google declared that only applications whose primary function necessitates a full-screen notification will by default be granted Full Screen Intent permission for Android 14 and above, and all other apps will have to request permission to use this permission.

New Covid variant evolution can be predicted by a novel model

0

Researchers claim to have created a model that forecasts the likely evolution of COVID-19-causing S-CoV-2 virus variations.

The model’s predictions of viral evolution, according to researchers from the University of Cologne in Germany and the Icahn School of Medicine in the US, can be used to help create vaccines that are as effective as possible against potential future variations.

Predictive modeling is a mathematical technique that examines patterns in a given set of input data to forecast future occurrences or results.

Which of the Covid variants can evade human immunity, proliferate throughout the population, and ultimately emerge as new major variants is predicted by the model.

It also predicts probable escape evolution pathways before new Covid variants appear, according to a study that was published in the journal Cell. In subsequent epidemic waves during the S-CoV-2 pandemic, a new variant superseded the old one.

But now, a number of Covid variations are in competition with one another and are circulating concurrently throughout the world.

Predictive models are then crucial in identifying the variants that are most likely to become dominant in the near future, according to the researchers.

On the one hand, internal alterations in new variants may improve transmissibility. However, they noted that immunity to earlier versions has been developed through prior infections or immunizations.

These variables are used by the new model to compute fitness differences between variants.The study demonstrates that the primary driver of S-CoV-2’s evolution is now human immunity.

“This implies that the virus will continue to generate new variants even after the pandemic is over, and we need timely data on them,” Cologne Institute for Biological Physics Professor Michael Lassig said.

Lassig, the study’s lead author, stated, “Our results show how important continued, internationally coordinated surveillance of the S-CoV-2 virus is.”

The World Health Organization receives recommendations on influenza vaccine selection from Lassig’s lab, which has been working on predictive analysis of the flu virus for years.

Snapchat hits 406m daily users, reports $368m net loss

0

The parent company of Snapchat, Snap, reported a net loss of $368 million for the quarter that ended on September 30. While this represents a slight improvement over the same period last year, the business is still expanding slowly.

The company announced late on Tuesday that Jerry Hunter, who was appointed chief operating officer just over a year ago, is retiring and will not be replaced.

Compared to the same period last year, Snap’s total revenue increased by 5% to almost $1.2 billion.

Snapchat has 406 million daily active users, a 12% year-over-year increase.

Executive Chairman of Snap, Evan Spiegel, stated, “We have evolved our go-to-market efforts to better serve our partners and drive customer success. We are focused on improving our advertising platform to drive higher return on investment for our advertising partners.”

When COO Hunter joined Snap seven years ago, he played a significant part in the development of the engineering and organizational structures of the company.

He will stay on board with Snap until July 1, 2024, in order to facilitate the transition.

Spiegel expressed her sincere gratitude to Jerry for his significant contributions during his lengthy tenure at Snap. “His efforts to strengthen our team, give back to the community, and enhance our advertising platform have helped lay the groundwork for our future growth.”

Over 200 million users have sent over 20 billion messages through Snap’s AI-powered chatbot, “My AI,” since its launch.

As compared to Q3 2022, there have been almost three times as many public Stories posted in the US on Snapchat by creators. The company stated, “We are committed to growing and diversifying our revenue growth.

In the December quarter, Snapchat expects to have between 410 and 412 million daily active users.

Microsoft CEO admits Giving Up on Windows phone

0

CEO and chairman of Microsoft Satya Nadella has stated that the company made a mistake by leaving the smartphone business, one that could have been handled more skillfully.

Microsoft initially announced in 2017 that it would no longer be developing new hardware or features for Windows 10 mobile devices, following years of difficulties selling Windows smartphones to consumers who preferred Apple’s iOS and Google’s Android operating systems (OS).

Users of Windows 10 Mobile stopped getting bug patches, security updates, and help options on December 10, 2019.

Nadella acknowledged that giving up on Windows Phone and mobile was a mistake in his most recent interview with Business Insider, which was published this week.

He was cited in the report as saying, “The decision I think a lot of people talk about — and one of the most difficult decisions I made when I became CEO — was our exit of what I’ll call the mobile phone as defined then.”

Nadella continued, “Looking back, I believe there were ways we could have made it work, maybe by redefining the category of computing between PCs, tablets, and phones.”

In 2014, Nadella succeeded former Microsoft CEO Steve Ballmer.

Microsoft slashed 7,800 jobs in 2015, mostly in the phone division, and wrote off $7.6 billion associated with its purchase of the Nokia phone division.

In an email to staff members, Nadella stated, “We are moving from a strategy to grow a standalone phone business to a strategy to grow and create a vibrant Windows ecosystem including our first-party device family.”

In 2017, Microsoft ultimately acknowledged the demise of Windows Phone.

Bill Gates, the former CEO and co-founder of the business, previously declared that Microsoft’s defeat by Android was his “greatest mistake ever.”

AI-Powered Playlist Art debuts on YouTube Music

0

Google-owned YouTube Music debuts AI-Powered Playlist Art. YouTube’s music app now offers a new feature that allows users to use generative Artificial Intelligence (AI) to create custom playlist art.

The new experimental feature was released by the company for English language users in the United States, with plans to expand its availability to other regions and languages in the future.

According to a blog post by YouTube, “We’ll be rolling out a new experimental feature that creates customized playlist art using generative AI for English language US users on YouTube Music starting today.”

According to the company, the feature will let users easily explore a variety of visual themes and options to create original cover art that captures the distinctiveness of your own playlists.

Additionally, in the upcoming months, the company will be releasing a new feature on the Home tab that will enable users to quickly browse and listen to the songs and artists they have been listening to a lot.

It’s easy to return to your current favorites by opening the YouTube Music app and finding your most-listened-to music content right at the top of the Home tab, according to YouTube.

In the meantime, YouTube has revealed fresh approaches to support journalists in producing dynamic news content and let users delve deeper into the newest news stories.

According to the platform, it is launching an engaging watch page for YouTube news articles. The news watch page will compile content from reliable sources for podcasts, live streams, video on demand, and shorts, enabling users to delve deeper and examine a variety of sources and viewpoints.

Future Medical Procedures Using Plant-Based Microbots

0

Plant-based smart materials have been developed by a research team at the University of Waterloo in Ontario. These materials will serve as the foundation for a new generation of tiny medical microrobots.

These robots may be used to perform minimally invasive medical procedures like biopsy and the transportation of cells and tissues. They can carry light and delicate cargo, like tissues or cells, to a target location by moving through flooded and cramped environments, like the human body.

The minuscule soft robots are non-toxic and biocompatible, with a maximum length of one centimeter. The research, which was published in the journal Nature Communications, states that the robots are constructed from sophisticated hydrogel composites that contain plant-based, sustainable cellulose nanoparticles.

Hamed Shahsavan, a professor in the chemical engineering department, is leading this research, which presents a comprehensive method for the creation, synthesis, manufacturing, and control of microrobots.

When the hydrogel used in this project is subjected to outside chemical stimulation, it changes shape.

The ability to program such shape-change, which is essential for the creation of functional soft robots, is made possible by researchers’ ability to orient cellulose nanoparticles at will.

“We utilize conventional soft matter such as hydrogels, liquid crystals, and colloids to present novel microrobots,” stated Shahsavan.

The advanced smart material’s self-healing feature is another distinctive feature that enables a broad range of robot shapes to be programmed.

In order to form different shapes for different procedures, researchers can simply cut the material and paste it back together without the need for glue or other adhesives.

A magnetism that makes it easier for soft robots to navigate the human body can be added to the material.

Scaling the robot to submillimeter scales is the next stage of this research.

X No More Free, Basic Features cost $1 now

0

The social media site X, formerly known as Twitter, said on Tuesday that it will launch a new membership plan in which basic functions will cost $1 year.

On the web version of the site, users will be charged for likes, reposts, quotes from other accounts, and bookmarking content under the new “Not A Bot” subscription.

X claims that the new subscription model is being implemented in order to combat spammers and bots. The price will vary based on the currency rate used in each nation.

Customers in the Philippines and New Zealand will be the first to use the new method, according to X.

This test has no effect on people who are currently using it. If they decide not to subscribe, new users may just view and read posts, watch videos, and follow accounts.

Elon Musk and the company have clashed about bots ever since Musk purchased the company last year. In July, X placed a cap on how many tweets users may read “to ensure the authenticity” of its user base.

X CEO Linda Yaccarino met with X’s lenders earlier this month, and according to a source cited by Reuters, she stated the business would test three tiers of its subscription service based on the amount of adverts presented to the user.

AI tool to identify people with abnormal heart rhythms

0

An algorithm based on artificial intelligence (AI) has been developed by researchers to identify atrial fibrillation, or irregular heart rhythm, in individuals who do not exhibit symptoms.

An irregular and frequently extremely fast heart rhythm known as atrial fibrillation (arrhythmia) can cause blood clots within the heart. It raises the risk of heart failure, stroke, and other heart-related issues.

The algorithm, which found hidden signals in common medical diagnostic tests, may help doctors better prevent strokes and other cardiovascular complications in patients with the most common type of heart rhythm disorder, according to the Smidt Heart Institute team at Cedars-Sinai Medical Center.

According to experts, around one in three patients with atrial fibrillation are unaware that they have the disorder.

The electrical impulses in the heart that control the flow of blood from the upper chambers to the lower chambers are disorganized in atrial fibrillation.

An ischemic stroke may result from blood pooling in the upper chambers and forming blood clots that go to the brain. Researchers used an artificial intelligence tool to analyse trends in ECG data in order to construct the algorithm.

A test that tracks the electrical impulses coming from the heart is called an electrocardiogram. Electrodes that measure cardiac electrical activity are placed to test patients’ bodies.

The software was trained to analyze ECG data gathered from January 1, 1987, to December 31, 2022. After studying over a million ECGs, the system correctly predicted that patients will get atrial fibrillation in less than 31 days.