Despite Big Privacy Claims, Meta, Apple, and Google Share User Data with Law Enforcement Upon Request

Pavel Durov, Telegram’s CEO, was recently arrested and now the users of Telegram are concerned whether the CEO will give their data to law enforcement or not. Telegram has always been involved in giving a platform to criminals who do a lot of things from drug trade to selling explicit videos and images of minors and adults. It is almost considered a dark web for dummy criminals. As Durov was arrested, many users came to know that the platform isn’t encrypted as it claimed to be, and now their data can easily be accessed by law enforcements.

Telegram had handed over some of its data to German authorities in January 2024 despite it claiming to protect users’ privacy. This makes us wonder if our data on different social media apps is as private as we think. So the answer is no about 70% to 80% of the time. Despite many big companies like Apple, Google and Meta campaigning to protect their users’ privacy, they would easily handover your data to authorities when requested.

Between January and June 2023, there were 270,000 law enforcement requests received by Meta for user data and nearly 77% of the data was disclosed to the authorities by the company. Google also disclosed 81% of the user data to law enforcement agencies in May 2023. Same goes for Apple as it disclosed 83% of user data in the first half of 2023 after 180,00 requests received, and 78% of the data in the second half of 2023.

These aren’t the only big tech companies to give user data to authorities as TikTok is also doing it. In 2023, 10,000 requests were received by TikTok for user data. There is a “Transparency Report” on TikTok's website which shows the data disclosed by country by them. In the USA, 71% of the information was disclosed, 68% in the UK and 57% in Australia.

It shows that it is pretty easy for law enforcements to get access to your data and the companies often hand it over to them without much resistance. The data law enforcement authorities get from these companies include names, addresses, contact information, emails, messages, videos, images and calendars and even contact lists. There is also a gag order given by authorities to these companies that they can only say that your data has been handed over to the authorities after six months to one year after the data has been disclosed.

The Guardian says that law enforcement authorities can obtain your data by “key search warrant” or “reverse search warrant” and also from “geofence warrant” or “reverse location warrant”. Itt means that your type of address or location at a specific time can become a part of the crime. So, Telegram users and users of other apps should know that if you are a part of some criminal activities, law enforcement authorities probably know.

Read next: Bombshell Leak Reveals Big Tech’s Marketing Partners Listen to Users’ Smartphone Microphones
by Arooj Ahmed via Digital Information World

YouTube Liaison Answers Most Asked Questions About YouTube Algorithm and SEO

There has always been talks about how rankings on YouTube are done and which factors and variables make rankings affected. YouTube Liaison in a recent interview answered some of the common questions YouTube creators ask about its SEO strategies. The first question that was asked was if the category of the YouTube video is important for the algorithm. The answer is that it isn’t that important and most of the time, YouTube doesn’t consider it. But still, content creators should try to choose a category that is the best aligned with their/viewer interest.

YouTube Liaison said that if you want to please the algorithm, please the audience and make videos which the audience loves and wants to watch. When asked if tags are important for YouTube videos, the answer was that they aren’t really important. It is better that you put efforts in your title and description than the tags. This question was followed by the question about hashtags and if they really matter. Still, the answer was that hashtags don’t matter that much unless you are hoping on a trend.

The YouTube Liason was also asked if consistency matters in the video performance or not. The answer is yes, it does matter but not in the way that you think. YouTube consistency matters in a way that you upload new videos everyday, at the same time and follow a strict schedule. This way your audience will be able to connect more with your content. Keywords in the video description also matter for the algorithm but try to not stuff them.

Image: DIW-Aigen

Another question was about why some videos take too much time to take off. The answer is again the audience. Sometimes your core audience likes the video but not the general audience. Sometimes, your core audience doesn’t find the video engaging but the general audience loves the video but it will take some time for the video to reach them.

Some other things that were found through this QnA was that low retention on embedded videos doesn’t matter to the algorithm and the algorithm also doesn’t look at your file name. The last question was if keeping different types of content like shorts, long form videos and lives on the same channel confuses the algorithm. If your audience is interested in all types of content, then all these content types are worth keeping on the same channel. But if you think that different types of audience likes different content, try making content on different channels.

You can watch the complete video here:



Read next: Survey Reveals Pinterest as Top Choice for User-Generated Content Marketing, Outpacing X
by Arooj Ahmed via Digital Information World

Pressure From EU Forces X To Abort Training AI Chatbot Grok With User Data

Elon Musk has just confirmed that his X app has withdrawn plans to train the popular AI chatbot Grok with user data. The news comes after facing scrutiny and pressure from EU officials.

The region is known for its tough regulatory laws. Last month, X suspended the move temporarily after it was legally prosecuted by the DPC in Ireland for the practice. No tech giant wants to engage in legal proceedings for this reason when the DPC is involved. Hence, it seems like X has gone with the permanent decision to abort any such data training plans.

The news was a prompt decision and this immediately resulted in all legal proceedings being conducted against the company. Remember, the DPC is the region’s top regulator, and seeing it raise concerns against X was never a good look for Musk and his executives at X.

The company has been giving its users quite some time now to opt out of the data training practice that it would employ for Grok. Moreover, it tried to defend the move by stating it was a default action that was set up for the public.

Under the GDPR, so many companies must get explicit consent before taking part in the practice. The DPC also issued officials with a notice to ensure what data processing really meant and how it was being used for AI training. This includes how personal data would benefit the firm.

After seeing X’s prompt action, we won’t be seeing it get scrutinized further for Grok and the DPC hopes Musk’s company and others can remain proactive and follow these regulations that remain in place throughout Europe.

Image: DIW-Aigen

Read next: Deep Fakes Pose Serious Risk to Scientific Research, Says Oxford's Sibel Erduran
by Dr. Hura Anwar via Digital Information World

Deep Fakes Pose Serious Risk to Scientific Research, Says Oxford's Sibel Erduran

Professor of Science Education and Director for Research at Oxford University, Sibel Erduran, wrote a paper about how altered content on the internet is affecting scientific studies and research. As artificial intelligence (AI) is getting common, deep fakes have also seen a rise. With the rise in technology, deep fakes are becoming more real. Erudan says that scientific research requires trust and that is not possible with the amount of altered information available on the internet.

In 2023, thousands of research papers were recalled due to fake content present in them. The professor says that scientific integrity is already under threat, and now deep fakes are adding another layer of threat to them. Deep fakes can also alter the videos of respected scientists and make them say wrong information which can become a big problem. Issues like climate change and public health are under the most threat.

Erudan says that if we make deep fake detection tools, it can save a bit of scientific integrity. On the other hand, deep fakes can give some opportunities for learning too. They can be beneficial to technological advancements and innovation. For instance, deep fakes can serve as realistic simulation for medical students to learn medical skills without compromising safety.

More than half of Americans have been fooled by deep fakes. According to a survey, 33% to 50% Americans couldn’t identify deep fakes. To keep ourselves safe from deep fakes, we should focus on deep learning and enhance our critical thinking skills. Misinformation detection tools and deep fake detection tools can help us combat fake content.

Image: DIW-Aigen

Read next: Despite Their Digital Expertise, Gen-Z Remains the Biggest Victims of Online Scams
by Arooj Ahmed via Digital Information World

Survey Reveals Pinterest as Top Choice for User-Generated Content Marketing, Outpacing X

According to Digiday+ Research, many digital marketers are thinking about making social media marketing strategies and budgets on platforms that rely on user-generated content. X is the biggest user-generated content (UGC) focused app and marketers have started making marketing strategies on this platform since 2022. But the new survey reveals that Pinterest, another UGC focused platform, is becoming a favorite of many marketers this year. Many retailers and marketers are budgeting more ad spend for Pinterest than for X.

Digiday+ found out that Pinterest is ranking the highest among other UGC focused platforms (X and Twitch). Pinterest also has the highest brand usage than X, which is a huge change from last year. 30% of the brand retailers have said that they have posted content on Pinterest last month while 21% said that they have posted brand content on X last month. Comparatively, 28% of the brand retailers and marketers said that they posted brand content on Pinterest last summer, while 35% said that about X. This shows that now marketers are interested about Pinterest more than X. On the other hand, Twitch is far behind these platforms with only 11% marketers saying that they have posted content on Twitch last month.

Looking at brands' ad spend on Pinterest last summer, 26% of brands said that they purchased advertising on Pinterest but this summer, only 21% of brands purchased advertising on the platform. There can be several reasons for this dip, but the prominent one is probably Pinterest’s push to grow its advertising last year. Many brands purchased ads last year because they were seeing how Pinterest and its teams were pushing advertising, but they soon saw that their purchased ads didn’t pay off on Pinterest. Even with this dip, Pinterest is still the top UGC focused platform in terms of ad spend. Both X and Twitch just saw 8% marketers purchasing ad spend this summer.

Digiday’s survey also found out that X is constantly declining in ad spend and usage. In summer 2022, 73% of marketers and brand retailers posted content on X which then declined to 35% in summer 2023 and soon fell to 21% in summer 2024. Similar is the case with purchasing advertisements on X. In summer 2022, 21% brands purchased ads while 14% did so in summer 2023. This year the ad spending on the platform has declined to just 8%.

Although Twitch is not that popular among brands, its ad spending has remained steady. 11% brands did marketing on Twitch this year but 9% the summer before. In summer 2022, 12% brands and marketers advertised on the platform. The ad spending on Twitch was 8% this year, 7% in 2023 and 6% in 2022.

Read next: The Social Media App Everyone’s Flocking To: Reddit’s Astonishing Rise in User Engagement!
by Arooj Ahmed via Digital Information World

Google Makes Makes Major Update To Chrome With Silent Integration Of Advanced Gemini AI Chabot In Address Bar

Search engine giant Google has just upgraded its Chrome browser with a major AI update that was rolled out discreetly.

Recently, Google integrated the Gemini chatbot into the address bar, starting from Chrome Canary version, which gave users access to some great AI advancements. It’s certainly a major step that would alter the way many carry out searches through Google Chrome browser .

Experts predict it would replace classic means of making search queries with the help of Generative AI in the replies. Now, you can directly chat with Gemini through the address bar.

The company gives Chrome users the chance to access the chatbot by typing @gemini right after their respective question in the browser’s address bar (just be sure you are an active user of Gemini and logged into your Google account). Therefore, users don’t need to waste time going to separate websites or adding apps to better engage with the audience for the best AI assistance.

This certainly makes AI a default means of browsing on Chrome as compared to before where classic search mechanisms were used.

Well, we won’t call the feature new as tests started in the middle of April and slowly launched into Chrome’s Canary as a beta. But after its successful integration, the company is launching this for all users to get wider adoption.

The arrival of this feature comes in Chrome’s general version and that marks Google’s goal of adding more AI to different core products. We also are aware of Gemini 1.5 Flash which gives more users access to the best AI technology through browsers directly.

While we agree it’s not as specialized as Gemini 1.5 Pro, this Flast variant does still give a lot of improvements to what its predecessors offered. Still, the level is not the same as Microsoft’s Copilot which does give users more contextual awareness when browsing. Hence, assistance on specific websites cannot be provided with the feature.

Just last month, Google spoke about the launch of several AI offerings for Chrome such as Google Lens to conduct visual searches. Similarly, it added a tab comparison for online e-shopping and better offerings for browsing history.

Now, this latest AI feature inside address bars is a major step towards the escalation of the entire AI movement by Google across its suite. As far as what the impact of these offerings would be, it could really raise AI accessibility for hundreds of millions of global users. It’s a widespread availability that accelerates AI tool adoption for daily tasks and makes productivity better. This means the common user can gain access to information better and quicker.

However, the latest rollout is not going unnoticed by critics who have raised important concerns. Some of these have to do with data privacy and the greater role of AI in our everyday lives. Amongst those, the biggest concern has to do with data collection and profiling of users.

It also means a major makeshift change for the business and tech world. The rollout signals a shifting landscape in software and how data is managed. This forces firms to reassess their tech stacks and how to better compete against AI-based platforms. Furthermore, it might have a big impact on the world of SEO and digital marketing.

More users will use AI browsing and that could alter their behaviors in terms of consuming information online. How business plans to optimize their presence and increase engagement will be affected.

When we look at this feature in the AI race against other arch-rivals, Google’s decision is a strategic one to ensure it’s the leader in browsers and AI. Chrome is the most popular with a massive user base. So equipping this with AI expands the company’s footprint in this division.

Read next: OpenAI Is Working On Eight More Voices For ChatGPT That Express Sounds More Naturally
by Dr. Hura Anwar via Digital Information World

OpenAI Is Working On Eight More Voices For ChatGPT That Express Sounds More Naturally

OpenAI is working on giving users a more seamless and natural experience while using the popular chat assistant.

If you’ve become bored of the constant or repetitive sounds that come out of the chatbot during use, here’s some great news. The tool is getting up to eight more natural sounds. This brings the total number of options to 12 for this AI endeavor.

Voice support has always been one of the chatbot’s most unique features and seeing this get a revamp is certainly going to be welcomed by users.

This news was first confirmed by Testing Catlog which thanks to reverse engineering shared the revelation of eight new voice options coming in a future launch. Interestingly, each voice has a specific name including Fathom, Harp, Orbit, Reef, Rainbow, Glimmer, Maple, Ridge, and Vale.

This would be up for grabs alongside the usual four options that we’ve been using since the start. Other than giving a distinct tone, the fresh offerings can also express non-verbal sounds. This adds a more realistic touch.

Furthermore, the new voices will be able to imitate animals and produce sounds so much more accurately. Common examples described so far are dog barks. Other than that, the latest array of sounds can give rise to more adjustments for tones depending on how text is formatted. Hence, if something is highlighted in bold, it could express a feeling with greater emotion than another in plain style. The same would go for italics.

For now, it’s not clear if and when the release will happen as OpenAI is yet to provide any confirmation on this front. All the reports are generated through reverse engineering so it’s still a work in progress.

Read next: Survey Reveals Top and Bottom Performers in Junior Employee Satisfaction: Salesforce and Google Lead, Amazon Trails
by Dr. Hura Anwar via Digital Information World