OnePlus could push bloatware, third-party apps on its premium smartphones

OnePlus could push bloatware, third-party apps on its premium smartphones


OnePlus is among the select smartphone brands that offer clear user interface experience in its smartphones. However, this might change soon. Reportedly, the company has started the soft-preload cycle with the OnePlus 12 and OnePlus Open, pushing third-party apps on the devices with the new OxygenOS update.


According to a report by Android Authority, OnePlus presents a “Review additional apps” page during the setup process on the OnePlus 12 running the latest Oxygen OS. In addition to the Google apps such as Google Drive, Google Home, YouTube and more, OnePlus has added a new section on this page where four third-party apps are pre-selected for installation. These apps include LinkedIn, Policybazaar, Block Blast and Candy Crush Saga.  While these apps can be deselected from the menu, it is difficult for a user to locate these pre-selected non-Google apps. The report stated that these apps also appear during the setup process on the company’s premium foldable smartphone- OnePlus Open.


The consumer technology news website stated that it has spotted evidence of references to other bloatware apps including a “Must Play” folder with games like Bubble Pop, Tile Match and more, within the Oxygen OS 14.0.0.610 firmware. This suggests that OnePlus is planning to add more apps to the list on its premium smartphones.


Responding to Android Authority, a OnePlus spokesperson said “The soft-preloads on the OnePlus 12 was an error made during testing and has been rectified as of 6 May. The OnePlus 12 does not come pre-loaded with any of these apps and will continue to remain light, fast and smooth.”


However, Android Authority noted that the setup process remained unchanged even after OnePlus said that the error was rectified.

First Published: May 09 2024 | 11:54 AM IST



Source link

Huawei's latest high-end phone uses more China-made parts, memory chip

Huawei's latest high-end phone uses more China-made parts, memory chip


Huawei (Photo: Bloomberg)


Huawei’s latest high-end phone features more Chinese suppliers, including a new flash memory storage chip and an improved chip processor, a teardown analysis showed, pointing to the progress China is making towards technology self-sufficiency.


Online tech repair company iFixit and consultancy TechSearch International examined the inside of Huawei Technologies’ Pura 70 Pro for Reuters, finding a NAND memory chip they said was likely packaged by the Chinese telecoms equipment maker’s in-house chip unit HiSilicon and several other components made by Chinese suppliers.


These findings have not been previously reported.


Huawei’s resurgence in the high-end smartphone market after four years of US sanctions is being widely watched by both rivals and US politicians as it has become a symbol of growing US-China trade frictions and China’s bid for technology self-sufficiency.


The firms also found that the Pura 70 phones run on an advanced processing chipset made by Huawei called the Kirin 9010 that is likely only a slightly improved version of the Chinese-made advanced chip used by Huawei’s Mate 60 series.


“While we cannot provide an exact percentage, we’d say the domestic component usage is high, and definitely higher than in the Mate 60,” said Shahram Mokhtari, iFixit’s lead teardown technician.


“This is about self-sufficiency, all of this, everything you see when you open up a smartphone and see whatever are made by Chinese manufacturers, this is all about self-sufficiency,” Mokhtari said.


Huawei declined to comment.


Huawei launched the Pura 70’s four smartphone models in late April and the series quickly sold out. Analysts say it will likely take more market share from iPhone manufacturer Apple, while policymakers in Washington are questioning the efficacy of US curbs on the telecoms equipment giant.


China-made Flash Memory Chip


Earlier analysis by teardown firms such as TechInsights of the Mate 60, launched in August last year, found the phone to be using DRAM and NAND memory chips made by South Korea’s SK Hynix. SK Hynix said at the time it no longer did business with Huawei and analysts said the chips likely came from stockpiles.


The Pura 70 still contains a DRAM chip made by SK Hynix, iFixit and TechSearch found, but the NAND flash memory chip was likely packaged by Huawei’s HiSilicon unit this time around and was made up of NAND dies each with a capacity of 1 terabit. This is comparable to products made by major flash memory producers such as SK Hynix, Kioxia and Micron.


However, the firms were unable to definitively identify the manufacturer of the wafer as the markings on the NAND die were unfamiliar, they added. But iFixit added that they believed that HiSilicon may have produced the memory controller as well.


“In our teardown our chip ID expert has identified it as a particular HiSilicon chip,” Mokhtari said.


SK Hynix reiterated that it was “strictly complying with the relevant policies since the restrictions against Huawei were announced and has also suspended any transactions with the company since then”.


Incremental improvements


IFixit and TechSearch’s analysis of the processor used by the Pura 70 Pro also suggests Huawei may have only made incremental improvements in its ability to produce an advanced chip with Chinese partners in the months since it launched the Mate 60 series.


The processor is similar to the one employed in the Mate 60 series that was produced for Huawei by Semiconductor Manufacturing International Corp (SMIC) using the Chinese chip foundry’s 7 nanometer (nm) N+2 manufacturing process, they said.


“This is significant because news of the 9000S on a 7nm node caused a bit of a panic last year when US lawmakers were confronted with the possibility that the sanctions imposed on Chinese chipmakers might not slow their technological progress after all,” iFixit said.


“The fact that the 9010 is still a 7nm process chip, and that it’s so close to the 9000S, might seem to suggest that Chinese chip manufacturing has indeed been slowed.”


Still, he cautioned against underestimating Huawei, saying that SMIC was still expected to make a leap to a 5nm manufacturing node before the end of the year.


SMIC did not respond to a request for comment.

(Only the headline and picture of this report may have been reworked by the Business Standard staff; the rest of the content is auto-generated from a syndicated feed.)

First Published: May 09 2024 | 9:52 AM IST



Source link

Best form of search will involve combination of narrative answers: Pichai

Best form of search will involve combination of narrative answers: Pichai



Emily Chang




Alphabet Inc. Chief Executive Officer Sundar Pichai says artificial intelligence has been a key focus of the Google parent since 2016, back when ChatGPT-maker OpenAI was in its infancy. After all, Google researchers invented the “T” in GPT (as in generative pre-trained transformer). It was a critical innovation that made conversational search using large language models possible.

 


Somehow though, Google missed the big chatbot moment and has been playing catchup ever since. But Pichai, who sat down for an exclusive interview with The Circuit with Emily Chang, doesn’t seem worried. “We weren’t the first company to do search. We weren’t the first company to do email. We weren’t the first company to build a browser,” he says. “So I view this AI as we are in the earliest possible stages.”

 


In other words, Pichai is playing the long game, and says Google—which dominates key real estate on the web—has plenty of time to win.

 


Google’s effort to reclaim the AI microphone has nevertheless experienced more than a few hiccups. When the company unveiled its Gemini image generator in February, users quickly found weaknesses. Requests for depictions of historical scenes yielded awkward images of Asian Nazis and Black US founding fathers. The company’s effort to ensure its AI systems didn’t perpetuate human biases had apparently backfired. 

 


“We got it wrong,” says Pichai, 51, who contends the incident was a case of good intentions gone awry. Google immediately shut down Gemini’s image generation feature for people, with Pichai ordering a complete rebuild. “From the ground up we are retraining these models, just to make sure we are also making the product better,” he says. “As soon as it’s ready, we will get it out to people.” He predicted the feature will be re-released in a few weeks.

 


Still, the future of search—and whether Google will continue to dominate that space—remains unclear. Next week, Pichai is to share his vision for the company’s future at Google I/O, the company’s annual developers conference. But in his interview with The Circuit, he gave us a preview.

 


Are we nearing the end of those “ten blue links,” like some pundits have predicted, as more conversational results from ChatGPT, Antrhopic’s Claude and other chatbots become more mainstream? Pichai says the best form of search will involve some combination of narrative answers and links to other websites to allow further exploration.

 


“My son is celiac, so we did a quick question to see whether something is gluten free,” Pichai says. Often a search for one query “leads to more things and then you want to explore more.” Meeting a variety of a searcher’s needs is what makes Google unique, he says.

 


Getting search right is essential to Google’s future, since ads placed among search results drive Alphabet’s $300 billion in annual revenue. “We’ve always found people want choices, including in commercial areas, and that’s a fundamental need,” Pichai explains. “We’ve been experimenting with ads and the data we see show that those fundamental principles will hold true.”

 


It’s also at the heart of a landmark antitrust lawsuit in which the US Department of Justice accuses Google of abusing its market power to illegally maintain a monopoly over online search and related advertising. A federal judge is expected to rule on the case later this year, and his decision could have far-reaching consequences for Alphabet’s business and beyond. 

 


But if Pichai is worried about Alphabet being broken up, he doesn’t show it. “People are trying to solve problems in their day-to-day lives,” he says. “A lot of our products integrate in a way that provides value for our users.” The way Google is approaching AI “drives innovation, adds choice in the market,” he says. “That’s how I think about it.”

 

In the meantime, there are plenty of hurdles Google will need to clear to arrive at its AI future. An increasing amount of AI-generated content is showing up on the internet, and all search engines will have to figure out how to track, categorize and surface it for users—or not. For example, last year Google’s algorithms inadvertently made an AI-generated “selfie” its top image for searches of “tank man,” the Chinese man who famously stood before tanks leaving Tiananmen Square in 1989.

Google took that image down from its Knowledge Graph and Knowledge Panels, but not before it momentarily gave people a new twist on history.  

 


“The challenge for everyone and the opportunity is: How do you have a notion of what’s objective and real in a world where there’s going to be a lot of synthetic content?” Pichai asks. “I think it’s part of what will define search in the next decade ahead.” 


Generative AI is developing so quickly that soon the large language models sponging up information on the internet will run out of content to feed on. That will lead to a situation where models are turning to AI-generated data to train on. 

 


Pichai says however that there are ways in which models being trained on synthetic data could lead to useful research breakthroughs. For instance, Google built up AlphaGo, its AI system that’s been trained to master the Japanese board game Go, in part by letting computer programs play with each other. “In the field, you call this self play,” Pichai says. “Over time, there’s this notion of can you have models create outputs for other models to learn? These are all research areas now.”

 


Amid all the strategic challenges Alphabet faces, Pichai is also confronting some skepticism from within the Google ranks.

Current and former employees have criticized his leadership style as too cautious and consensus-driven, citing that as a reason Google ceded the lead in AI to ChatGPT, at least initially. 

 


“The reality I think is quite different,” Pichai argues. “I think the larger the company is, you are making fewer consequential decisions, but they need to be clear and you have to point the whole company to that.” He says it’s important to build consensus because “that’s what allows you to have maximum impact behind those decisions.”

 


Pichai has recently taken several steps to streamline the business to focus more on AI—including layoffs. Alphabet has gone through multiple rounds of cuts in divisions including hardware, engineering and the Google Assistant team.

 


Last month, Google also fired dozens of engineers who protested the company’s cloud contract with the Israeli government, in what Pichai describes as an unacceptable disruption of daily business. “It has nothing to do with the matter or the topic they’re discussing. It’s about the conduct of how they went about it,” he says. “I view, particularly in this moment with AI, the opportunity we have ahead of us is immense, but it needs a real focus on our mission.”

 


Microsoft Corp., with its big investments in AI startups like OpenAI, Inflection and Mistral AI, has emerged as Alphabet’s biggest rival in the most frenzied tech cycle since the dotcom boom. The two are vying to win the AI race, realizing the technology holds the key to the future of search—which is essential to the future of AI.

 


Microsoft CEO Satya Nadella has had some fighting words for Google, too. During the US antitrust trial, Nadella testified that Google’s exclusive, multibillion-dollar deals with the likes of Apple Inc., as the default on smartphones and browsers, have essentially locked out Microsoft’s Bing. Google has been able to make its search engine better because, as the default, it gets more user queries, which in turn lead to better search results, Nadella argued.

 


Pichai meanwhile says he’s trying to keep focused and not “play to someone else’s dance music.”

 


“People tend to focus in this micro moment, but it is so small in the context of what’s ahead,” Pichai says. “When I look at the opportunities ahead, across everything we do, I put a lot of chips, at least from my perspective, on Google.”

 


This episode of The Circuit With Emily Chang debuts Wednesday, May 8, at 6 p.m. in New York on Bloomberg Television, on the Bloomberg app, Bloomberg.com and the Bloomberg Originals YouTube Channel. Check out The Circuit podcast for extended conversations.

 


–With assistance from Lauren Ellis, Julia Love and Davey Alba.

 


©2024 Bloomberg L.P.



Source link

Red Hat defines AI strategy; CEO says impact of open source visible in AI

Red Hat defines AI strategy; CEO says impact of open source visible in AI


Red Hat has also made a slew of announcements at the flagship event, alongside a slew of partnerships and collaborations. Photo: Intel


The impact of open source ecosystem in the realm of AI is becoming visible and the momentum is set to accelerate, Red Hat CEO Matt Hicks has said as the IBM-owned enterprise open source solutions provider outlined its strategy and push on AI.


At his keynote address at the Red Hat Summit, Hicks predicted that Artificial Intelligence (AI) won’t be built by a single vendor, rather it will be defined by flexibility of choices.


He observed that innovation has moved at a fast clip, with advancements unfolding every month, week and hour.


Open source and academia are at the centre of innovation, he asserted.


According to him, the staggering rate and pace of academic advancement has found the “perfect intersection” in Open Source and AI models.


Today, the capabilities in AI are being developed by some of the best academic researchers, with Open Source helping unlock potential, he pointed out.


“The most interesting thing that happened to AI over last year has been explosion of choices, what used to be only achievable in trillion perimetre models is being now being replicated in models orders of magnitude smaller… capabilities that just a year ago were coupled to high end, fairly-exotic hardware can now run on a laptop,” he said.


At the same time, training techniques that once ran in hundreds of millions of dollars are being replicated for a fraction of the costs, said the top boss of Red Hat.


Red Hat, which was acquired by IBM in a $34 billion mega deal in 2018, is laying bold bets on AI.

 


This year, the Red Hat Summit focuses on open source and AI, and “incredible impact” when the two are combined, Hicks said.


“We are now seeing impact of broadening Open Source ecosystem in the world of AI and this is not going to slowdown as models get smaller, as training gets cheaper, as capabilities grow, the knowledge of this tech will expand,” Hicks said.

He added, “The breakthrough from MIT today will be expanded on by IITs in India tomorrow… researchers will use this as a channel for their next concept and in turn they will equip developers to build their next idea.”

AI, he said, won’t be built by a single vendor, nor is it going to revolve around a single monilithic model.


“… your choice where to run AI will be everywhere and it is going to be based on Open Source,” he said.


Red Hat has also made a slew of announcements at the flagship event, alongside a slew of partnerships and collaborations.


The provider of open source software solutions has announced the launch of Red Hat Enterprise Linux AI (RHEL AI), a foundation model platform that enables users to develop, test and deploy generative AI (GenAI) models more seamlessly.


Red Hat further announced that it has infused Generative AI across hybrid cloud portfolio with Red Hat Lightspeed.


“Red Hat Lightspeed will expand generative AI capabilities to Red Hat’s industry-leading Linux and hybrid cloud application platforms, helping to make complex computing tasks more efficient and accessible across skill levels,” the company said in a release.


Red Hat Inc has also announced a collaboration with Intel to power enterprise AI usages on Red Hat OpenShift AI.


Together, the two companies will facilitate the delivery of artificial intelligence (AI) solutions on Intel AI products for model development and training, model serving, management and monitoring more seamlessly across a hybrid cloud infrastructure.

(Only the headline and picture of this report may have been reworked by the Business Standard staff; the rest of the content is auto-generated from a syndicated feed.)

First Published: May 08 2024 | 8:14 PM IST



Source link

HC asks Centre to respond to plea on non-regulation of deepfake technology

HC asks Centre to respond to plea on non-regulation of deepfake technology


The plea said in view of these threats, there is a pressing need for strict enforcement and proactive action to mitigate potential harm associated with their misuse.Photo: Wikimedia Commons


The Delhi High Court on Wednesday sought the Centre’s response on a PIL by veteran journalist Rajat Sharma against non-regulation of deepfake technology in the country and seeking directions to block public access to applications and software enabling creation of such content.


A bench of Acting Chief Justice Manmohan and Justice Manmeet PS Arora issued notice to the Union Ministry of Electronics and Information Technology on the petition.


Deepfake technology facilitates creation of realistic videos, audio recordings and images that can manipulate and mislead viewers by superimposing the likeness of one person onto another, altering their words and actions, thereby presenting a false narrative or spreading misinformation.


The bench said it was a major problem and sought to know from the central government if it was willing to act on the issue.


Political parties are complaining about this as well. You are not taking any action, the court said.


Rajat Sharma, the Chairman and Editor-in-Chief of Independent News Service Private Limited (INDIA TV), said in the public interest litigation (PIL) that proliferation of deepfake technology poses significant threat to various aspects of society, including misinformation and disinformation campaign, and undermines the integrity of public discourse and the democratic process.


The PIL said there is a threat of potential use of this technology in fraud, identity theft and blackmail, harm to individual reputation, privacy and security, erosion of trust in media and public institutions and violation of intellectual property rights and privacy rights.


The plea said in view of these threats, there is a pressing need for strict enforcement and proactive action to mitigate potential harm associated with their misuse.


The absence of adequate regulation and safeguards against the misuse of deepfake technology poses a grave risk to the fundamental rights guaranteed under the Constitution of India, including the right to freedom of expression, the right to privacy, and the right to a fair trial, the petition said.


It said it is imperative for the government to establish regulatory frameworks to define and classify deepfakes and AI-generated content and prohibit the creation, distribution and dissemination of deepfakes for malicious purposes.


The failure of the government to regulate and provide effective redressal to victims of deepfakes and AI-generated content effectively not only undermines the rights and safety of its citizens but also jeopardizes the integrity of democratic institutions and societal trust, it said.


The plea said the Centre had made a statement of its intent to formulate regulation for dealing with deepfakes and synthetic content in November 2023, but nothing of the sort has seen the light of the day so far.


The petitioner mentioned in the plea that certain unscrupulous people were maintaining social media accounts and uploading fake videos featuring his image and his Artificial Intelligence (AI)-generated voice to sell or endorse various products such as purported medication for diabetes and fat loss.


The PIL sought a direction to the Centre to identify and block public access to the applications, software, platforms and websites enabling the creation of deepfakes.


It also sought a direction to appoint a dedicated nodal officer for receiving complaints regarding deepfakes and acting upon them within 12 hours and within 6 hours in the case of a complaint received regarding content featuring a public figure.


The plea said that the government be asked to issue a directive to all social media intermediaries to initiate immediate action to take down deepfakes upon receipt of a complaint from the person concerned.

(Only the headline and picture of this report may have been reworked by the Business Standard staff; the rest of the content is auto-generated from a syndicated feed.)

First Published: May 08 2024 | 7:24 PM IST



Source link

NPU: What is it and why is it gaining importance in semiconductor industry

NPU: What is it and why is it gaining importance in semiconductor industry


Neural Engine on the Apple M4 chip

Apple debuted the M4 chip with the iPad Pro on May 7. While Apple stated significant improvements in chip’s overall performance, the most prominent change in the M4 is the 16-core Neural Engine, which is Apple’s term for Neural Processing Unit (NPU). Apple called the M4 an “outrageously powerful chip for AI.” It is the improvements in the NPU that is backing Apple’s claim of the M4 being significantly more powerful than any other chip powering an AI PC. So, what exactly is NPU and why is it gaining prominence in the chip industry? Let us find out:


What is NPU (Neural Processing Unit)


An NPU, or a Neural Processing Unit, is a dedicated processor designed specifically for accelerating neural network processes. A neural network is essentially a type of machine learning algorithm that mimics the human brain for processing data. Therefore, the NPU is highly capable for handling machine learning operations that form the basis for AI-related tasks, such as speech recognition, natural language processing, photo or video editing processes like object detection, and more.


In most consumer-facing gadgets such as smartphones, laptops and tablets, the NPU is integrated within the main processor, adopting a System-on-Chip (SoC) configuration. However, for data centres, the NPU might be an entirely discrete processor, separate from any other processing unit such as the central processing unit (CPU) or the Graphics processing unit (GPU).


How is NPU different from CPU and GPU


CPUs employ a sequential computing method, issuing one instruction at a time, with subsequent instructions awaiting the completion of their predecessors. In contrast, NPU harnesses parallel computing to simultaneously execute numerous calculations. This parallel computing approach results in swifter and more efficient processing. That said, CPUs are good at sequential computing, executing one process at a time, but running AI tasks requires the processor to execute multiple calculations and processes simultaneously.


This is where the Graphic Processing Units, or the GPUs, come in. These have parallel computing capabilities and have integrated circuits embedded to carry out AI workloads, but they are generally meant for handling other processes, such as graphic rendering and resolution upscaling. It essentially makes a case for NPUs, which simply replicate those circuits and make it dedicatedly work on carrying out machine learning operations. This allows the AI workload processing to be more efficient and less power consuming.


GPUs are still used in the initial development and refinement of AI algorithms, while NPUs later on takes the mantle by running those refined language models on the consumer’s device.


NPU and on-device AI


Large language models (LLMs) are too big to run on-device and service providers generally take up the processing to the cloud to offer AI features based on their language models. However, recently big technology companies have released small language models such as Google’s Gemma, Microsoft’s Phi-3 and Apple’s OpenELM, indicating a trend towards small scaled AI models, capable of running entirely on-device. As on-device AI models gain more prominence, the role of NPUs get even more crucial as they are the one deploying AI-powered applications on the hardware.

First Published: May 08 2024 | 5:21 PM IST



Source link

YouTube
Instagram
WhatsApp