At Google I/O on Wednesday, Google said that all Gmail users in the U.S. will soon be able to discover if their email address has been found on the dark web. The dark web report security feature will roll out over the coming weeks, and will be expanded to select international markets. BleepingComputer reports: Once enabled, it will allow Gmail users to scan the dark web for their email addresses and take action to protect their data based on guidance provided by Google. For instance, they'll be advised to turn on two-step authentication to protect their Google accounts from hijacking attempts. Google will also regularly notify Gmail users to check if their email has been linked to any data breaches that ended up on underground cybercrime forums.

"Dark web report started rolling out in March 2023 to members across all Google One plans in the United States, providing a simple way to get notified when their personal information was discovered on the dark web. "Google One's dark web report helps you scan the dark web for your personal info -- like your name, address, email, phone number and Social Security number -- and will notify you if it's found," said Google One Director of Product Management Esteban Kozak in March when the feature was first announced. The company says all the personal info added to the profile can be deleted from the monitoring profile or by removing the profile in the dark web report settings.

An anonymous reader shares an excerpt from a collaborative investigation between Motherboard, lavialibera, and IrpiMedia: Mafioso Bartolo Bruzzaniti needed everyone to do their job just right. First, the Colombian suppliers would hide a massive amount of cocaine inside bananas at the port city of Turbo, Colombia. That shipping container would then be transported across the ocean to Catania, in Sicily, Italy. A corrupt port worker on the mafia's payroll would wave the shipment through and had advised the group how to package the drugs. This was so the cocaine could remain undetected even if the worker was forced to scan the shipment. Another group of on-the-ground mafiosos would then unload the cocaine outside of the port.

In March 2021, Bruzzaniti, an alleged member of the infamous 'Ndrangheta mafia group and who says Milan belongs to him "by right," asked his brother Antonio to go fetch something else crucial to the traffickers' success. "Go right now," Bruzzaniti wrote in a text message later produced in court records. "It's needed urgently." Investigators know what Bruzzaniti said because European authorities had penetrated an encrypted phone network called Sky and harvested around a billion of the users' messages. These phones are the technological backbone of organized crime around the world.

The thing Antonio needed to urgently fetch was a phone from a different encrypted phone network, one that the authorities appear to have not compromised and which the mafia have been using as part of their operations. To that phone, a contact sent one half of the shipping container's serial number. A reporting collaboration between Motherboard, lavialibera, and IrpiMedia has identified that encrypted phone as being run by a company called No. 1 Business Communication (No. 1 BC). The investigation has found members of the mafia and other organized crime groups turning to No. 1 BC as authorities cracked down on other platforms. The collaboration has identified multiple key players in No. 1 BC's development, sales, and legal structure. "Take the bc1 right away," Bruzzaniti wrote in another text, referring to the No. 1 BC phone.

A ransomware intrusion on hardware manufacturer Micro-Star International, better known as MSI, is stoking concerns of devastating supply chain attacks that could inject malicious updates that have been signed with company signing keys that are trusted by a huge base of end-user devices, a researcher said. From a report: "It's kind of like a doomsday scenario where it's very hard to update the devices simultaneously, and they stay for a while not up to date and will use the old key for authentication," Alex Matrosov, CEO, head of research, and founder of security firm Binarly, said in an interview. "It's very hard to solve, and I don't think MSI has any backup solution to actually block the leaked keys."

The intrusion came to light in April when, as first reported by Bleeping Computer, the extortion portal of the Money Message ransomware group listed MSI as a new victim and published screenshots purporting to show folders containing private encryption keys, source code, and other data. A day later, MSI issued a terse advisory saying that it had "suffered a cyberattack on part of its information systems." The advisory urged customers to get updates from the MSI website only. It made no mention of leaked keys. Since then, Matrosov has analyzed data that was released on the Money Message site on the dark web. To his alarm, included in the trove were two private encryption keys. The first is the signing key that digitally signs MSI firmware updates to cryptographically prove that they are legitimate ones from MSI rather than a malicious impostor from a threat actor. This raises the possibility that the leaked key could push out updates that would infect a computer's most nether regions without triggering a warning. To make matters worse, Matrosov said, MSI doesn't have an automated patching process the way Dell, HP, and many larger hardware makers do. Consequently, MSI doesn't provide the same kind of key revocation capabilities.

An anonymous reader shares a report: YouTube Premium subscribership grew to a record 80 million users in 2022, and Google responded by announcing it would be investing more into its subscription offerings in 2023. What we didn't realize at the time was how that could mean handicapping its free offerings to get more people to pay for its services. When watching videos yesterday, one Redditor encountered a popup informing them that "Ad blockers are not allowed on YouTube." The message offered a button to "Allow YouTube ads" in the person's ad blocking software and went on to explain that ads make the service free for billions of users and that YouTube Premium offers an ad-free experience. It even provided a button to easily sign up for a YouTube Premium membership.

Amazon CTO Werner Vogels, writes in a blog post: Software architectures are not like the architectures of bridges and houses. After a bridge is constructed, it is hard, if not impossible, to change the way it was built. Software is quite different, once we are running our software, we may get insights about our workloads that we did not have when it was designed. And, if we had realized this at the start, and we chose an evolvable architecture, we could change components without impacting the customer experience. My rule of thumb has been that with every order of magnitude of growth you should revisit your architecture, and determine whether it can still support the next order level of growth.

A great example can be found in two insightful blog posts written by Prime Video's engineering teams. The first describes how Thursday Night Football live streaming is built around a distributed workflow architecture. The second is a recent post that dives into the architecture of their stream monitoring tool, and how their experience and analysis drove them to implement it as a monolithic architecture. There is no one-size-fits-all. We always urge our engineers to find the best solution, and no particular architectural style is mandated. If you hire the best engineers, you should trust them to make the best decisions.

I always urge builders to consider the evolution of their systems over time and make sure the foundation is such that you can change and expand them with the minimum number of dependencies. Event-driven architectures (EDA) and microservices are a good match for that. However, if there are a set of services that always contribute to the response, have the exact same scaling and performance requirements, same security vectors, and most importantly, are managed by a single team, it is a worthwhile effort to see if combining them simplifies your architecture.

Evolvable architectures are something that we've taken to heart at Amazon from the very start. Re-evaluating and re-architecting our systems to meet the ever-increasing demands of our customers. You can go all the way back to 1998, when a group of senior engineers penned the Distributed Computing Manifesto, which put the wheels in motion to move Amazon from a monolith to a service-oriented architecture. In the decades since, things have continued to evolve, as we moved to microservices, then microservices on shared infrastructure, and as I spoke about at re:Invent, EDA.

An anonymous reader quotes a report from Ars Technica: The US Justice Department has seized the domains of 13 DDoS-for hire services as part of an ongoing initiative for combatting the Internet menace. The providers of these illicit services platforms describe them as "booter" or "stressor" services that allow site admins to test the robustness and stability of their infrastructure. Almost, if not all, are patronized by people out to exact revenge on sites they don't like or to further extortion, bribes, or other forms of graft. The international law enforcement initiative is known as Operation PowerOFF. In December, federal authorities seized another 48 domains. Ten of them returned with new domains, many that closely resembled their previous names.

"Ten of the 13 domains seized today are reincarnations of services that were seized during a prior sweep in December, which targeted 48 top booter services," the Justice Department said. "For example, one of the domains seized this week -- cyberstress.org -- appears to be the same service operated under the domain cyberstress.us, which was seized in December. While many of the previously disrupted booter services have not returned, today's action reflects law enforcement's commitment to targeting those operators who have chosen to continue their criminal activities." According to a seizure warrant (PDF) filed in federal court, the FBI used live accounts available through the services to take down sites with high-capacity bandwidth that were under FBI control. "The FBI tested each of services associated with the SUBJECT DOMAINS, meaning that agents or other personnel visited each of the websites and either used previous login information or registered a new account on the service to conduct attacks," FBI Special Agent Elliott Peterson wrote in the affidavit. "I believe that each of the SUBJECT DOMAINS is being used to facilitate the commission of attacks against unwitting victims to prevent the victims from accessing the Internet, to disconnect the victim from or degrade communication with established Internet connections, or to cause other similar damage."

A vulnerability in the "Advanced Custom Fields" plugin for WordPress is putting more than two million users at risk of cyberattacks, warns Patchstack researcher Rafie Muhammad. The Register reports: A warning from Patchstack about the flaw claimed there are more than two million active installs of the Advanced Custom Fields and Advanced Custom Fields Pro versions of the plugins, which are used to give site operators greater control of their content and data, such as edit screens and custom field data. Patchstack researcher Rafie Muhammad uncovered the vulnerability on February 5, and reported it to Advanced Custom Fields' vendor Delicious Brains, which took over the software last year from developer Elliot Condon. On May 5, a month after a patched version of the plugins was released by Delicious Brains, Patchstack published details of the flaw. It's recommended users update their plugin to at least version 6.1.6.

The flaw, tracked as CVE-2023-30777 and with a CVSS score of 6.1 out of 10 in severity, leaves sites vulnerable to reflected XSS attacks, which involve miscreants injecting malicious code into webpages. The code is then "reflected" back and executed within the browser of a visitor. Essentially, it allows someone to run JavaScript within another person's view of a page, allowing the attacker to do things like steal information from the page, perform actions as the user, and so on. That's a big problem if the visitor is a logged-in administrative user, as their account could be hijacked to take over the website.

"This vulnerability allows any unauthenticated user [to steal] sensitive information to, in this case, privilege escalation on the WordPress site by tricking the privileged user to visit the crafted URL path," Patchstack wrote in its report. The outfit added that "this vulnerability could be triggered on a default installation or configuration of Advanced Custom Fields plugin. The XSS also could only be triggered from logged-in users that have access to the Advanced Custom Fields plugin."

An anonymous reader quotes a report from TechCrunch: NextGen Healthcare, a U.S.-based provider of electronic health record software, admitted that hackers breached its systems and stole the personal data of more than 1 million patients. In a data breach notification filed with the Maine attorney general's office, NextGen Healthcare confirmed that hackers accessed the personal data of 1.05 million patients, including approximately 4,000 Maine residents. In a letter sent to those affected, NextGen Healthcare said that hackers stole patients' names, dates of birth, addresses and Social Security numbers. "Importantly, our investigation has revealed no evidence of any access or impact to any of your health or medical records or any health or medical data," the company added. TechCrunch asked NextGen Healthcare whether it has the means, such as logs, to determine what data was exfiltrated, but company spokesperson Tami Andrade declined to answer.

In its filing with Maine's AG, NextGen Healthcare said it was alerted to suspicious activity on March 30, and later determined that hackers had access to its systems between March 29 and April 14, 2023. The notification says that the attackers gained access to its NextGen Office system -- a cloud-based EHR and practice management solution -- using client credentials that "appear to have been stolen from other sources or incidents unrelated to NextGen." "When we learned of the incident, we took steps to investigate and remediate, including working together with leading outside cybersecurity experts and notifying law enforcement," Andrade told TechCrunch in a statement. "The individuals known to be impacted by this incident were notified on April 28, 2023, and we have offered them 24 months of free fraud detection and identity theft protection." NextGen was also the victim of a ransomware attack in January this year, adds TechCrunch. The stolen data, including employee names, addresses, phone numbers and passport scans, appears to be available on the dark web.

An anonymous reader quotes a report from The Register: This year's DEF CON AI Village has invited hackers to show up, dive in, and find bugs and biases in large language models (LLMs) built by OpenAI, Google, Anthropic, and others. The collaborative event, which AI Village organizers describe as "the largest red teaming exercise ever for any group of AI models," will host "thousands" of people, including "hundreds of students from overlooked institutions and communities," all of whom will be tasked with finding flaws in LLMs that power today's chat bots and generative AI. Think: traditional bugs in code, but also problems more specific to machine learning, such as bias, hallucinations, and jailbreaks -- all of which ethical and security professionals are now having to grapple with as these technologies scale. DEF CON is set to run from August 10 to 13 this year in Las Vegas, USA.

For those participating in the red teaming this summer, the AI Village will provide laptops and timed access to LLMs from various vendors. Currently this includes models from Anthropic, Google, Hugging Face, Nvidia, OpenAI, and Stability. The village people's announcement also mentions this is "with participation from Microsoft," so perhaps hackers will get a go at Bing. We're asked for clarification about this. Red teams will also have access to an evaluation platform developed by Scale AI. There will be a capture-the-flag-style point system to promote the testing of "a wide range of harms," according to the AI Village. Whoever gets the most points wins a high-end Nvidia GPU. The event is also supported by the White House Office of Science, Technology, and Policy; America's National Science Foundation's Computer and Information Science and Engineering (CISE) Directorate; and the Congressional AI Caucus.

Apple failed to revive a long-running copyright lawsuit against cybersecurity firm Corellium over its software that simulates the iPhone's iOS operating systems, letting security researchers to identify flaws in the software. From a report: The US Court of Appeals for the Eleventh Circuit on Monday ruled that Corellium's CORSEC simulator is protected by copyright law's fair use doctrine, which allows the duplication of copyrighted work under certain circumstances. Apple argued that Corellium's software was "wholesale copying and reproduction" of iOS and served as a market substitute for its own security research products. Corellium countered that its copying of Apple's computer code and app icons was only for the purposes of security research and was sufficiently "transformative" under the fair use standard. The three-judge panel largely agreed with Corellium, finding that CORSEC "furthers scientific progress by allowing security research into important operating systems" and that iOS "is functional operating software that falls outside copyright's core."

An anonymous reader shared this report from NBC News: Alexej Savreux, a 34-year-old in Kansas City, says he's done all kinds of work over the years. He's made fast-food sandwiches. He's been a custodian and a junk-hauler. And he's done technical sound work for live theater.

These days, though, his work is less hands-on: He's an artificial intelligence trainer.

Savreux is part of a hidden army of contract workers who have been doing the behind-the-scenes labor of teaching AI systems how to analyze data so they can generate the kinds of text and images that have wowed the people using newly popular products like ChatGPT. To improve the accuracy of AI, he has labeled photos and made predictions about what text the apps should generate next.

The pay: $15 an hour and up, with no benefits... He credits the AI gig work — along with a previous job at the sandwich chain Jimmy John's — with helping to pull him out of homelessness.
"Their feedback fills an urgent and endless need for the company and its AI competitors: providing streams of sentences, labels and other information that serve as training data," the article explains: "A lot of the discourse around AI is very congratulatory," said Sonam Jindal, the program lead for AI, labor and the economy at the Partnership on AI, a nonprofit based in San Francisco that promotes research and education around artificial intelligence. "But we're missing a big part of the story: that this is still hugely reliant on a large human workforce," she said...

A spike in demand has arrived, and some AI contract workers are asking for more. In Nairobi, Kenya, more than 150 people who've worked on AI for Facebook, TikTok and ChatGPT voted Monday to form a union, citing low pay and the mental toll of the work, Time magazine reported... Time magazine reported in January that OpenAI relied on low-wage Kenyan laborers to label text that included hate speech or sexually abusive language so that its apps could do better at recognizing toxic content on their own. OpenAI has hired about 1,000 remote contractors in places such as Eastern Europe and Latin America to label data or train company software on computer engineering tasks, the online news outlet Semafor reported in January...

A spokesperson for OpenAI said no one was available to answer questions about its use of AI contractors.

ChatGPT creator OpenAI is now valued at nearly $30 billion, reports Fortune — adding that CEO Sam Altman "still thinks startups are most effective when employees work together in an office." The idea of fully remote work becoming the norm has come and gone, he said this week at a fireside chat in San Francisco organized by the fintech company Stripe. "I think definitely one of the tech industry's worst mistakes in a long time was that everybody could go full remote forever, and startups didn't need to be together in person and, you know, there was going to be no loss of creativity," he told attendees. "I would say that the experiment on that is over, and the technology is not yet good enough that people can be full remote forever, particularly on startups."

He isn't alone in his assessment. Many CEOs have been demanding that remote employees spend more time in the office, among them Bob Iger at Disney, Howard Schultz at Starbucks, and Robert Thomson at News Corp. During the pandemic, remote work or a hybrid work schedule was the only option for many office workers — and many grew to prefer it to being in the office every workday.

"I do not believe in remote work for startups," Keith Rabois, a general partner at venture capital firm Founders Fund, told The Logan Bartlett Show last week, adding that neither he nor his firm would invest in a venture based on it. Younger workers, he noted, "learn by osmosis" in a way that requires in-person interaction, and supervisors discover hidden talent by watching them...

Altman said, "I feel pretty strongly that startups need a lot of in-person time, and the more fragile and nuanced and uncertain a set of ideas are, the more time you need together in person."

An anonymous reader shared this report from Fortune: For eight years now, as millennials have entered their thirties and forties, also known as "homebuying age," Bank of America has surveyed over 1,000 members of the generation once a year for its Home Work series. And for 2023's edition... older millennials (age 31-41) are almost three times as likely to move into a house than an apartment, the survey found...

Migration patterns during the pandemic have clearly established that most homebuyers have wanted to flee big cities, with some "zoomtowns" such as Boise benefiting in particular. But the survey reveals something even more drastic. In a section called "suburban nation," BofA reveals that 43% to 45% of millennials — of every age — expect to buy a house in the suburbs. "We expect the ability to work from home to remain an incentive for young families to seek out more remote suburban and rural markets where housing may be more affordable," wrote the BofA team led by research analyst Elizabeth Suzuki. And remote work is still robust, they added.

Millennials are also looking toward the suburbs for wealth-building. A majority (two-thirds) of them believe that they'll buy a home in the next two years, citing a return on investment as the number one reason for purchasing. The interest is pervasive across the generation, and maybe means that the suburb is in for a new and better revival. And a 2021 study from Pew Research Center found that one in five adults preferred city life, compared to one quarter of adults in 2018...

Millennials reported to BoA that the pandemic increased their likelihood of buying a home...

US authorities have seized another major Z-Library domain but still haven't been able to wipe the pirate book site off the Internet. From a report: Z-Library claims to offer over 13 million books, up from 11 million since US authorities launched their first major operation against Z-Library late last year. "Unfortunately, one of our primary login domains was seized today," Z-Library wrote in a Wednesday message on its Telegram account. "Therefore, we recommend using the domain singlelogin[dot]re to log in to your account, as well as to register. Please share this domain with others." In November, US authorities charged Russian nationals Anton Napolsky and Valeriia Ermakova with criminal copyright infringement, wire fraud, and money laundering for allegedly operating Z-Library. The US said at the time that it seized 250 "interrelated web domains" run by Z-Library and that Napolsky and Ermakova were arrested in Argentina at the request of the US government. Other people continue to operate Z-Library, which remained available on the Tor network and returned to the clearnet in February with a new strategy of assigning personal, secret URLs to each user. Z-Library directed users to singlelogin[dot]me, where they could sign in with their login credentials and receive a unique URL to access the entire pirate library.

A judge sentenced Joe Sullivan, the former chief security officer at Uber, to three years' probation and 200 hours of community service on Thursday for covering up a 2016 cyberattack from authorities and obstructing a federal investigation. From a report: Sullivan's case is likely the first time a security executive has faced criminal charges for mishandling a data breach, and the response to Sullivan's case has split the cybersecurity community. In October, a jury found Sullivan guilty of obstructing an active FTC investigation into Uber's security practices and concealing a 2016 data breach that affected 50 million riders and drivers. Uber paid the hackers $100,000 to not release any stolen data and keep the attack quiet. Sullivan and his team routed the payment through the company's bug bounty program, which good-faith security researchers usually use to report flaws. The hack wasn't publicly disclosed until 2017, shortly after Dara Khosrowshahi stepped into the CEO role.

Khosrowshahi fired Sullivan in 2017, telling the jury last fall that he thought the decision to conceal the breach was "the wrong decision." Sullivan then joined Cloudflare as its chief security officer in 2018, and he stayed there until July 2022 when he stepped down to prepare for his trial. "If I have a similar case tomorrow, even if the defendant had the character of Pope Francis, they would be going to prison," Judge William Orrick said during the sentencing on Thursday. "When you go out and talk to your friends, to your CISOs, you tell them that you got a break not because of what you did, not even because of who you are, but because this was just such an unusual one-off," Orrick added.

An anonymous reader quotes a report from TechCrunch: The City of Dallas in Texas has confirmed a ransomware attack has downed key services, including 911 dispatch systems. City officials confirmed on Wednesday that a number of the city's servers had "been compromised with ransomware," causing widespread service outages. The Dallas Police Department (DPD) website is currently offline. The City of Dallas website displays a message stating that "the City is experiencing a service outage and is working to restore services," and the city wrote on a page that contains updates about the incident that all courts were closed on Wednesday and would be closed again on Thursday.

DPD spokesperson Melinda Gutierrez confirmed to TechCrunch that the outage has also impacted Computer Aided Dispatch, or "CAD" systems, which are used by dispatchers and 911 operators to prioritize and record incident calls. Local media reported that this has forced 911 call takers to manually write down instructions for responding officers. "There is no effect to 911 calls at this time, and they continue to be dispatched for service," Gutierrez added. "The outage is not affecting police response."

Printers on the City of Dallas network reportedly began printing out ransom notes on Wednesday morning. As per a copy the note, the Royal ransomware gang has claimed responsibility for the attack, and a URL included on the note directed to a contact form on Royal's dark web victims site. The note said critical data was encrypted, and threatened to publish it online if a ransom demand is not met. The City of Dallas has not yet been listed on Royal's dark web leak site and it's not yet known what types of data has been stolen. City officials have not responded to TechCrunch's questions. The full impact of the ransomware attack remains unknown. In a statement, the city said it was "actively working to isolate the ransomware to prevent its spread, to remove the ransomware from infected servers, and to restore any services currently impacted. The City is currently working to assess the complete impact, but at this time, the impact on the delivery of City services to its residents is limited."

An anonymous reader shares a report: We recently covered a study by Secure Data Recovery, an HDD, SSD, and RAID data recovery company, of 2,007 defective hard disk drives it received. It found the average time before failure among those drives to be 2 years and 10 months. That seemed like a short life span, but considering the limited sample size and analysis in Secure Data Recovery's report, there was room for skepticism. Today, Backblaze, a backup and cloud storage company with a reputation for detailed HDD and SSD failure analysis, followed up Secure Data Recovery's report with its own research using a much larger data set. Among the 17,155 failed HDDs Backblaze examined, the average age at which the drives failed was 2 years and 6 months.

Backblaze arrived at this age by examining all of its failed drives and their respective power-on hours. The company recorded each drive's failure date, model, serial number, capacity, failure, and SMART raw value. The 17,155 drives examined include 72 different models and does not include failed boot drives, drives that had no SMART raw attribute data, or drives with out-of-bounds data. If Backblaze only looked at drives that it didn't use in its data centers anymore, there would be 3,379 drives across 35 models, and the average age of failure would be a bit longer at 2 years and 7 months. Backblaze said its results thus far "are consistent" with Secure Data Recovery's March findings. This is despite Backblaze currently using HDDs that are older than 2 years and 7 months.

Dallas is experiencing IT and police communication outages following a ransomware attack on municipal systems. From a report: City officials said in a statement Wednesday that hackers had compromised "a number of servers" and that they were working with vendors to try to control the spread of malware. With a ransomware attack, hackers lock up victims' data or knock services offline, then demand an extortion payment. "We have been having a system shutdown for the past two days now," a Dallas 311 operator told Bloomberg News. "We are very limited in what we are able to access internally right now."

The Dallas police department's website was inaccessible when Bloomberg News tried to visit the page Thursday. The hack also affected the police dispatch system and resulted in the closure of local courts on Wednesday, TechCrunch reported. "We appreciate your patience during this time," the city website stated. The "Royal" ransomware gang claimed responsibility for the attack. The group sent city officials a note that included a link to communicate with the hackers and discuss a payment, however the size of the ransom demand was not immediately clear.

Discord is making "big changes" to how identities work on the platform, a move that will force you to change your username. From a report: Up until now, the company has appended four-digit tags to identities as a way to distinguish people with the same username. However, the new system will give everyone a unique username, much like Twitter, Instagram and other services. "The whole point of these changes is that we want to make it a lot easier for you and all the new users coming to Discord to connect and hang out with friends," co-founder Stanislav Vishnevskiy wrote in a blog post. "We know that your username and identity are important, and we understand that some of you may not like this change and disagree with it." The original aim with the four-digit tags was to allow you to choose any username you wanted, but it has now become "technical debt," according to Discord. The company said that the usernames are "too complicated or obscure" for people to remember.

Microsoft is making its Bing GPT-4 chatbot available to everyone today, no more waitlist necessary. From a report: All you need to do is sign in to the new Bing or Edge with your Microsoft account, and you'll now access the open preview version that's powered by GPT-4. Microsoft is also massively upgrading Bing Chat with lots of new features and even plug-in support. Microsoft is now adding more smart features to Bing Chat, including image and video results, new Bing and Edge Actions feature, persistent chat and history, and plug-in support. The plug-in support will be the key addition for developers and for the future of Bing Chat.