Tag: Science & Technology

With Microsoft hyping the Windows 11 AI-powered Recall feature, earlier this week a security researcher who previously worked for the tech giant called it a ‘security disaster.’ The feature takes a screenshot of everything you do on your PC and makes all of your past actions searchable. This may look great at first, but a security expert claims it stores data including passwords and card numbers in a plain text SQLite database.
Now, a developer named Alexander Hagenah has published a Python script that “copies the databases and screenshots and then parses the database for potentially interesting artifacts.” This means threat actors can automate the process of extracting passwords and other sensitive information and get away with it. Since the tool can be used to look for terms like “password” from the database and takes little to no time and does not break any encryption, it might allow hackers to steal all your information in mere seconds.
While Microsoft has repeatedly claimed that all Recall data is private and cannot be accessed by anyone other than the user, several security exploits that allow unauthorised users to get access to the database how come to light. The tech giant also claimed that threat actors will need physical access to your machine if they want to steal your passwords but malware might soon evolve to evade detection techniques.
For those not in the loop, the upcoming Windows 11 Recall will be limited to Copilot Plus PCs coming later this month, but the same security expert who called it a ‘security disaster’ claims that he was able to use it on a PC without an NPU, which Microsoft says is a pre-requisite for Recall to work.

Google has introduced a slew of features to its AI research and writing assistant NotebookLM. The upgraded version of the tool is powered by Google’s most capable model Gemini 1.5 Pro. With the latest upgrade, the tool will now be available across 200 countries and territories around the world.
“Our goal from the beginning with NotebookLM has been to create a tool to help you understand and explore complex material, make new connections from information, and get to your first draft faster. You can upload sources — your research notes, interview transcripts, corporate documents — and instantly NotebookLM becomes an expert in the material that matters most to you,” said Steven Johnson and Raiza Martin from Google Labs, in a blog post introducing the new upgrades.
What’s new in NotebookLM?
With the latest upgrade, NotebookLM now supports Google Slides and web URLs as sources along with Google Docs, PDFs, and text files. When it comes to inline citations, users will now be taken directly to supporting passages in their sources. This makes it easy to fact-check the AI response or dive deeper into the original text. Besides, the Notebook Guide gives you a high-level understanding of sources by converting them into useful formats like FAQs, Briefing Docs, or Study Guides. Owing to Gemini 1.5 Pro’s native multimodal capabilities, users can now ask questions about images, charts, and diagrams in Slides or Docs. NotebookLM will also include citations to images as supporting evidence whenever relevant.
When asked about the accuracy and reliability of citations, Martin said that NotebookLM generates citations for all its answers and it always cites the sources for every piece of information provided, ensuring that the answers are based on the sources one uploads.
“For instance, if you upload a source that contains information A, B, or C, any related questions you ask will pull answers directly from that source. To verify the credibility of the sources used, NotebookLM constantly monitors the quality of the answers. It performs checks to ensure that the answers are grounded in the uploaded sources. Additionally, quality is closely monitored whenever new features or upgrades are released or when new models are introduced to ensure there are no losses in quality,” Raiza told indianexpress.com.
When it comes to privacy and security of the source’s data and processes, NotebookLM ensures this by adhering to Google’s standard data protection practices. “Notably, NotebookLM does not train on any user data, ensuring that your data remains private. This privacy guarantee is particularly important for users who wish to use NotebookLM with sensitive work or school documents. Safeguards are in place to protect user information, maintaining confidentiality and security at all times,” Raiza said.
Source: The Indian Express

In the modern world of science and technology, we spend most time charging our electric appliances. But what if we tell you that you will now be able to charge your dead laptop or smartphone in just a minute? Well, the technology is here. Recently, an Indian-origin researcher has revealed a novel technology that can help in charging a laptop and phone in one minute. Moreover, using the same way, you can charge your electric car in 10 minutes only. Let’s see what this technology is.
How is it possible?
Ankur Gupta, an assistant professor of chemical and biological engineering at the US-based University of Colorado Boulder and his team of researchers discovered this new technology. In a study, published in the journal Proceedings of the National Academy of Sciences, researchers explained how tiny charged particles, called ions, move within a complex network of minuscule pores. The breakthrough could lead to the development of more efficient energy storage devices, such as supercapacitors, said Gupta, an assistant professor of chemical and biological engineering.
Gupta said, “Given the critical role of energy in the future of the planet, I felt inspired to apply my chemical engineering knowledge to advancing energy storage devices. It felt like the topic was somewhat unexplored and, as such, the perfect opportunity.”
Gupta explained that several chemical engineering techniques are used to study flow in porous materials such as oil reservoirs and water filtration, but they have not been fully utilised in some energy storage systems. The study stated that this discovery is not only for storing energy in vehicles and electronic devices but also for power grids, where fluctuating energy demand requires efficient storage to avoid waste during periods of low demand and to ensure rapid supply during high demand. Supercapacitors, energy storage devices that rely on ion accumulation in their pores, have rapid charging times and longer life spans compared to batteries. The researcher said that the primary appeal of supercapacitors is their speed. “So how can we make their charging and release of energy faster? By the more efficient movement of ions,” said Gupta.

Monster sunspot, designated AR3664, responsible for the May 10 solar storm that triggered the strongest auroras on Earth in over five centuries is back. After disappearing from Earth’s view, the sunspot bigger than the planet has returned. The gargantuan sunspot announced its arrival with a massive solar flare and the image of the explosion was captured by amateur astronomer Michael Karrer.
The explosion was so big that it caused a deep shortwave radio blackout over East Asia and hurled a bright coronal mass ejection (CME) into space. A CME is one of the most powerful explosions from the Sun that travels at millions of kilometres per hour carrying plasma and material across the vacuum of space.
In the early weeks of May this year, Earth was slammed by a series of seven solar storms over the weekend that triggered strong auroras in several parts of the world.
Travelling at a staggering speed of 48 lakh kilometres per hour, the coronal mass ejections barrelled through space to slam into Earth. Starting May 10, the intense activity kicked off a stunning display of auroras in several parts of the world including India, the rarest of rare occurrences.
The US National Oceanic and Atmospheric Administration issued a rare severe geomagnetic storm warning when a solar outburst reached Earth on Friday afternoon, hours sooner than anticipated. The effects of the Northern Lights, which were prominently on display in Britain, lasted through the weekend and are possibly to go into the week.The auroras were so bright that it was visible even in India. Hanle in Ladakh witnessed a bright red flash in the skies as the charged particles from the Sun struck into Earth’s magnetic field.As the solar storm struck Earth, space agencies across the world scrambled to protect assets and astronauts in space. American space agency, Nasa, India’s Isro and SpaceX all had to adjust and course-correct their satellites to evade the stress.
With the sunspot returning to view, more activity and explosions are expected. Source: India Today

Truecaller is collaborating with Microsoft to significantly upgrade its existing artificial intelligence (AI)-powered AI Assistant. As a part of this partnership, the caller identification app will leverage the Microsoft Azure AI Speech feature called Personal Voice. This will be integrated with the AI Assistant to give it a replica voice of the user when the AI screens a call or responds on behalf of the user. The feature is being rolled out starting today, and it will be available in a select few countries at first. The company has confirmed the feature will be expanded to more countries soon.
Making the announcement via a community post, Truecaller said, “With the addition of Microsoft’s Personal Voice, users of the Truecaller Assistant can now create a completely digital version of their own voice to use inside the Assistant.” Notably, the AI Assistant was first introduced by the company in September 2022. The voice-based chatbot offers multiple functionalities such as screening calls, responding to calls, and even taking messages on behalf of the user. It can also record calls that can be listened to later, similar to a voicemail.
While this is a useful feature, Truecaller also offered a limited number of voices that could be used by the AI Assistant. With this, when someone calls, they would hear one of these chosen voices before the call is picked by the user. However, hearing a stranger’s voice had the possibility of catching the caller off-guard. The app maker is solving this issue by introducing Personal Voice which will let people use their own voice for the AI Assistant.
This feature has started rolling out to Truecaller Premium users in the US, Canada, Australia, South Africa, India, Sweden, and Chile. It will be expanded to other regions soon, the company stated.
How to set up Personal Voice in AI Assistant
– One thing to note is that the feature is only available on Truecaller Premium. If you want to use it, you’ll have to purchase the subscription.
– To see the feature, you need to be on the latest version of the app. Ensure to update it.
– Once done, open the app and go to Settings.
– Go to Assistant settings.
– Go to Set Up Personal Voice and follow the instructions.
– You will have to record your voice for about a minute and speak out the words shown on the screen, after which your AI-generated voice replication process will be complete.

The Biden administration is poised to open up a new front in its effort to safeguard U.S. AI from China and Russia with preliminary plans to place guardrails around the most advanced AI models, Reuters has reported.
Government and private sector researchers worry U.S. adversaries could use the models, which mine vast amounts of text and images to summarize information and generate content, to wage aggressive cyber attacks or even create potent biological weapons.
DEEPFAKES AND MISINFORMATION
Deepfakes – realistic yet fabricated videos created by AI algorithms trained on copious online footage – are surfacing on social media, blurring fact and fiction in the polarized world of U.S. politics. While such synthetic media has been around for several years, it’s been turbocharged over the past year by a slew of new “generative AI” tools such as Midjourney that make it cheap and easy to create convincing deepfakes.
Image creation tools powered by artificial intelligence from companies including OpenAI and Microsoft, can be used to produce photos that could promote election or voting-related disinformation, despite each having policies against creating misleading content, researchers said in a report in March.
Some disinformation campaigns simply harness the ability of AI to mimic real news articles as a means of disseminating false information.
While major social media platforms like Facebook, Twitter, and YouTube have made efforts to prohibit and remove deepfakes, their effectiveness at policing such content varies.
For example, last year, a Chinese government-controlled news site using a generative AI platform pushed a previously circulated false claim that the United States was running a lab in Kazakhstan to create biological weapons for use against China, the Department of Homeland Security (DHS) said in its 2024 homeland threat assessment.
National Security Advisor Jake Sullivan, speaking at an AI event in Washington on Wednesday, said the problem has no easy solutions because it combines the capacity of AI with “the intent of state, non-state actors, to use disinformation at scale, to disrupt democracies, to advance propaganda, to shape perception in the world.” “Right now the offense is beating the defense big time,” he said.
BIOWEAPONS
The American intelligence community, think tanks and academics are increasingly concerned about risks posed by foreign bad actors gaining access to advanced AI capabilities. Researchers at Gryphon Scientific and Rand Corporation noted that advanced AI models can provide information that could help create biological weapons.
Gryphon studied how large language models (LLM) – computer programs that draw from massive amounts of text to generate responses to queries – could be used by hostile actors to cause harm in the domain of life sciences and found they “can provide information that could aid a malicious actor in creating a biological weapon by providing useful, accurate and detailed information across every step in this pathway.”
They found, for example, that an LLM could provide post-doctoral level knowledge to trouble-shoot problems when working with a pandemic-capable virus.
Rand research showed that LLMs could help in the planning and execution of a biological attack. They found an LLM could for example suggest aerosol delivery methods for botulinum toxin. Source: Reuters

Three of the Wall Street’s heavyweight technology firms have reported better-than-expected sales at their cloud computing units in recent days, as interest in artificial intelligence drives a rebound in spending by corporate customers.
Growth in the $270 billion cloud infrastructure market, a cash engine for Amazon.com, Microsoft and Alphabet, gives the clearest sign yet that AI investment is bearing fruit after investors drove those stocks to record highs, thanks to optimism about the emerging technology.
Many big customers have started spending again on cloud computing after pausing last year to cut costs, executives and analysts said.
Amazon, the last of the trio to report on Tuesday, said its cloud computing arm AWS grew 17% in the January-to-March period, above Wall Street’s 15% growth estimate, and hit a $100 billion annual run-rate for the first time.
Performance was consistent at Microsoft’s Azure and Google Cloud, which grew above expectations at 31% and 28%, respectively, in the first three months of the year.
“Looking across AWS, Microsoft Azure and Google Cloud, it is clear that two things are happening simultaneously – AI is contributing to growth, but also the rest of cloud spending is accelerating,” said D.A. Davidson & Co analyst Gil Luria.
For several years cloud infrastructure providers enjoyed growth rates as high as 60% and demand shot up during the COVID-19 pandemic as more businesses moved online. However, firms had to realign expectations last year as customers pulled back spending in an increasingly challenging business environment.
The industry has been at the forefront of adopting AI and customers have begun to buy the new functionality at a rapid pace, executives said.
“The number of Azure AI customers continues to grow and average spend continues to increase,” Microsoft CEO Satya Nadella said on the company earnings call, adding that more than 65% of the Fortune 500 companies were Azure OpenAI Service customers.
AI services contributed 7 percentage-points in growth to Azure, up from 6 percentage points in the Oct-Dec quarter.
More than 60% of funded generative AI startups and nearly 90% of genAI unicorns were using Google Cloud, Alphabet CEO Sundar Pichai said on his company’s earnings call last week.
“There is an inevitable and continuous migration of workloads to the cloud and consolidation of IT spending going towards large platforms, including the hyperscalers,” said RBC Capital Markets analyst Rishi Jaluria.
Hyperscalers are cloud providers with a large network of data centres and a wide range of services and are often preferred for end-to-end workload support. Source: Reuters