I spent a day on the dark web - these are the terrifying things I saw
Usually, our imaginations cook up scenarios that are a lot more dramatic than reality. But when it comes to this hidden corner of the internet that can only be reached using special software, it's pretty spot on.
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Daily Mail
an hour ago
- Daily Mail
The popular apps that are SPYING on you: Cybersecurity experts issue urgent warning over 'data hungry' apps that can access your location, microphone and data
They're some of the biggest apps in the world, used by hundreds of millions of people every day. But according to a new investigation, 'data hungry' smartphone apps like Facebook and Instagram ask for 'shocking' levels of access to your personal data. Experts at consumer champion Which? investigated 20 popular apps across social media, online shopping, fitness and smart home categories. They found all of them ask for 'risky' permissions such as access to your location, microphone, and files on your device – even when they don't need to. The experts urge people to be more careful about what exactly we agree to when we download an app and mindlessly agree to permissions. We could be compromising our privacy when we hastily tap 'agree'. 'Millions of us rely on apps each day to help with everything from keeping on top of our health and fitness to doing online shopping,' said Harry Rose, editor of Which? 'While many of these apps appear to be free to use, our research has shown how users are in fact paying with their data – often in scarily vast quantities.' Which? researchers worked with experts at cybersecurity firm Hexiosec to assess the privacy and security features of 20 popular apps on an Android handset. The list included some of the biggest names in social media (including WhatsApp, Facebook, Instagram, TikTok), online shopping (Amazon, AliExpress) the smart home (Samsung Smart Things, Ring Doorbell) and fitness (Strava). Combined, the 20 apps have been downloaded over 28 billion times worldwide – meaning the average UK adult is likely to have several of them on their phone at any given time. If someone were to have all 20 downloaded on their device, collectively they would grant a staggering 882 permissions – potentially giving access to huge amounts of an individual's personal data. Overall, the team found Chinese app Xiaomi Home asked for a total of 91 permissions – more than any other app in the study – five of which are described as 'risky'. Risky permissions include those that access your microphone, can read files on your device, or see your precise location (usually referred to as 'fine location'). Such data is a valuable commodity and may allow firms to target users with 'uncannily accurate adverts'. Samsung's Smart Things app asked for 82 permissions (of which eight are risky), followed by Facebook (69 permissions, six risky) and WhatsApp (66 permissions, six risky). Overall, Xiaomi asked for a total of 91 permissions - more than any other app in the study - five of which described as 'risky' Xiaomi Home was also one of two apps (alongside AliExpress) to send data to China, including to suspected advertising networks – although this was flagged in the privacy policy by both. Ali Express requested six risky permissions such as precise location, access to microphones and reading files on the device. AliExpress also bombarded users with a deluge of marketing emails after download (30 over the course of a month) but the researchers did not see any specific permission request from AliExpress to do so. Temu, another Chinese-owned online marketplace, also gave a heavy push to sign up to email marketing – which many users could easily agree to without realising, the experts reasoned. Among social media apps, Facebook was 'the most keen for user data' as it wanted the highest number of permissions (69 in total, six of which risky), followed by WhatsApp (66 altogether, six of which risky). TikTok, meanwhile, asked for 41 permissions, including three risky ones, including the ability to record audio and view files on the device, while YouTube asked for 47 permissions, four of which were 'risky'. Overall, 16 of the 20 apps requested a permission that allows apps to create windows on top of other apps – effectively creating pop-ups on your phone, even if you opted out of the app sending notifications. Seven also wanted a permission that allows an app to start operating when you open your phone even if you haven't yet interacted with it. In some cases there are clear uses for risky permissions – for example the likes of WhatsApp or Ring Doorbell may need microphone access in order to carry out certain functions. But other examples the need for risky permissions was less clear cut, according to Which? For example, four apps – AliExpress, Facebook, WhatsApp and Strava – requested permission to see what other apps recently used or currently running. The researchers stress that the investigation was conducted on an Android phone and that permissions may vary on Apple iOS devices. But we should all be more careful of tapping "yes" to permissions while mentally on 'autopilot' without really being aware of what we're agreeing to, Mr Rose said. 'Our research underscores why it's so important to check what you're agreeing to when you download a new app,' he added. The full findings can be read on the Which? website. In response to the findings, Meta (which owns WhatsApp, Facebook and Instagram) said none of its apps 'run the microphone in the background or have any access to it without user involvement'. Meta also said that users must 'explicitly approve' in their operating system for the app to access the microphone for the first time. A Samsung spokesperson said: 'All our apps, including SmartThings, are designed to comply with UK data protection laws and relevant guidance from the Information Commissioner's Office (ICO).' Meanwhile, TikTok said that privacy and security are 'built into every product' it makes. It added: TikTok 'collects information that users choose to provide, along with data that supports things like app functionality, security, and overall user experience'. Strava said that risky permission it takes, such as precise location, allow it to 'provide the very service that our users are requesting'. It said that it has 'implemented appropriate guardrails' around how data is 'collected, shared, processed, and used'. Amazon said that device permissions are to provide 'helpful features', such as 'the ability to visualise products in their home with their device's camera or search for products using text-to-speech'. It added: 'We also give customers clear control over personalised advertising by requesting consent when they visit our UK store and providing options to opt out or adjust preferences at any time.' AliExpress claimed that the precise location permission is not used in the UK, and the microphone permission requires user consent. It added: 'We strive to create a platform where consumers can shop with confidence, knowing that their data is safeguarded in accordance with the law and our strict privacy policy. We welcome the findings from Which? as an opportunity to redouble our efforts in this area.' Ring said that it doesn't 'use cookies or trackers on the Ring app for advertising' and all permission as used to 'provide user-facing features'. It added: 'We design our products and services to protect our customers' privacy and security, and to put our customers in control of their experience. We never sell their personal data, and we never stop working to keep their information safe.' A Temu spokesperson said precise location permission is 'used to support completing an address based on GPS location' but it is not used in the UK market, adding that it 'handles user data in accordance with local and international regulations and in line with leading industry practices'. Google (representing YouTube), Xiaomi, Impulse and MyFitnessPal did not respond to requests for comment.
Reuters
2 hours ago
- Reuters
Google and OpenAI's AI models win milestone gold at global math competition
July 21 (Reuters) - Alphabet's (GOOGL.O), opens new tab Google and OpenAI said their artificial-intelligence models won gold medals at a global mathematics competition, signaling a breakthrough in math capabilities in the race to build powerful systems that can rival human intelligence. The results marked the first time that AI systems crossed the gold-medal scoring threshold at the International Mathematical Olympiad for high-school students. Both companies' models solved five out of six problems, achieving the result using general-purpose "reasoning" models that processed mathematical concepts using natural language, in contrast to the previous approaches used by AI firms. The achievement suggests AI is less than a year away from being used by mathematicians to crack unsolved research problems at the frontier of the field, according to Junehyuk Jung, a math professor at Brown University and visiting researcher in Google's DeepMind AI unit. "I think the moment we can solve hard reasoning problems in natural language will enable the potential for collaboration between AI and mathematicians," Jung told Reuters. OpenAI's breakthrough was achieved with a new experimental model centered on massively scaling up "test-time compute." This was done by both allowing the model to "think" for longer periods and deploying parallel computing power to run numerous lines of reasoning simultaneously, according to Noam Brown, researcher at OpenAI. Brown declined to say how much in computing power it cost OpenAI, but called it "very expensive." To OpenAI researchers, it is another clear sign that AI models can command extensive reasoning capabilities that could expand into other areas beyond math. The optimism is shared by Google researchers, who believe AI models' capabilities can apply to research quandaries in other fields such as physics, said Jung, who won an IMO gold medal as a student in 2003. Of the 630 students participating in the 66th IMO on the Sunshine Coast in Queensland, Australia, 67 contestants, or about 11%, achieved gold-medal scores. Google's DeepMind AI unit last year achieved a silver medal score using AI systems specialized for math. This year, Google used a general-purpose model called Gemini Deep Think, a version of which was previously unveiled at its annual developer conference in May. Unlike previous AI attempts that relied on formal languages and lengthy computation, Google's approach this year operated entirely in natural language and solved the problems within the official 4.5-hour time limit, the company said in a blog post. OpenAI, which has its own set of reasoning models, similarly built an experimental version for the competition, according to a post by researcher Alexander Wei on social media platform X. He noted that the company does not plan to release anything with this level of math capability for several months. This year marked the first time the competition coordinated officially with some AI developers, who have for years used prominent math competitions like IMO to test model capabilities. IMO judges certified the results of those companies, including Google, and asked them to publish results on July 28. "We respected the IMO Board's original request that all AI labs share their results only after the official results had been verified by independent experts and the students had rightly received the acclamation they deserved," Google DeepMind CEO Demis Hassabis said on X on Monday. OpenAI, which published its results on Saturday and first claimed gold-medal status, said in an interview that it had permission from an IMO board member to do so after the closing ceremony on Saturday. The competition on Monday allowed cooperating companies to publish results, Gregor Dolinar, president of IMO's board, told Reuters.
Reuters
3 hours ago
- Reuters
NXP second-quarter revenue falls 6%
July 21 (Reuters) - Chipmaker NXP Semiconductors (NXPI.O), opens new tab posted a 6% drop in second-quarter revenue on Monday, led by weakness in communications and infrastructure segment amid broader market softness. NXP's revenue for the second quarter fell 6.4% to $2.93 billion, although it still narrowly beat analyst expectations of $2.90 billion, according to data compiled by LSEG. NXP's chips are used for high-speed digital processing utilized in sectors such as automotive, manufacturing, telecommunications and the Internet of Things (IoT). The shares of the company fell 5% in trading after the bell. Revenue from its communication and infrastructure segment fell 27% to $320 million in the quarter. Industrial and IoT revenue fell 11%, while the automotive segment was flat. For the third quarter, the company expects revenue to be between $3.05 billion and $3.25 billion, the midpoint of which is above analysts' estimates of $3.07 billion.
