Researchers Used Example of Nuclear Fission to Explain How a Rumor Spreads in a Society

Researchers from Shandong Normal University in China, led by Wenrong Zheng published a paper which talks about how a rumor doesn't go viral, instead gets nuclear. The researchers gave an example of how an infection spreads to make people understand the spread of information or misinformation. Just like people act as vectors for the spread of viruses, they are also a vector in spreading lies across masses.

It is a fact that lies travel faster than truth because people are more interested in listening to rumors than boring truths. The researchers say that the infectious disease model isn't that helpful in understanding the spread of lies. This is because diseases do not propagate actively but rumors do. The infectious disease model also just talks about the result of disease, that is, death but not other factors.

So what other natural processes can researchers use to study the spread of rumors? Many researchers have used a number of processes but the researchers of this study used nuclear fission to explain the phenomenon. Nuclear fission is the splitting of a nucleus into two or more nuclei by collision of uranium atom and neutron which produces large amounts of heat and radiation. The researchers named the rumors as neutrons, uranium atoms are named as receivers of rumors and barriers to fission are named as active propagation thresholds. The energy the fission produces is compared to the society’s reaction to online rumors.

When a neutron (rumor) collides with uranium atoms (people), the uranium produces energy according to the impact of collision. So, if a rumor is of high energy, people are going to get activated quickly and the spread will be larger too. Is the heat being produced by the activation of rumor getting utilized, that is, through user activity? Even though people are categorized as atoms, they still possess human qualities like how much someone is resistant to rumors, how educated a person hearing rumors is, how quickly someone can identify misinformation?

The overall heat produced by rumors is used to study how society is treating a rumor. The propagation can also be studied by using this nuclear model. It is an interesting way to look at how rumors can spread and how society is going to react to it. It can also help us understand which types of rumors society reacts quickly too and which types of rumors do not leave much impact after fission.

Image: DIW-Aigen

Read next: New Research Shows Tech Community has Extremely Polarized Thoughts About AI and Its Implications
by Arooj Ahmed via Digital Information World

Reddit CEO Justifies Blocking Microsoft From Crawling Its Site As Google Attains More Data

Last week, we heard about tech giant Reddit being accused of favoring Google while others such as Microsoft were blocked from crawling the platform.

Many questioned the behavior, forcing the company’s CEO to finally come forward and reveal the real reason why Bing was on the block list. And no surprises here as money is involved.

While most tech experts assumed an exclusive deal was signed behind close doors that gave Google an edge, the reality is far from that. CEO Steve Huffman explained how it’s got to do with the mere fact that Microsoft just won’t play ball.

It refuses to pay while Google is the exact opposite. This leaves the app with no choice but to prevent anyone crawling its site for free.

The deal that is in place with Google is worth a staggering $60M where the search giant can make use of Reddit data to improve its AI systems. Microsoft isn’t being left out, the CEO adds. It’s just that it refuses to negotiate and that’s why not a lot can be done there.

Seeing the drama and chaos unfold with all of these tech giants has been interesting. No deal or settlement was reached between Microsoft and Reddit, despite numerous discussions. Reddit accuses some other AI firms of reaching agreements that puts the company at a loss while there were quite a few unwilling to make promises that they could abide by.

Microsoft is saying no to negotiations because it feels Reddit does not even deserve a dime for that data. While the software giant refuses to go into the details on why it is being so stern, it did explain how it would accept and respect the painstaking task of being blocked from scrapping data from its page.

We feel that Microsoft is literally handing Google the driving seat position by ignoring Reddit’s demands. It’s just giving more people the chance to use Google than Bing, not to menton how the former’s marketshare is already hitting peak.

One thing that is interesting is how Microsoft striking a deal with Reddit is going to give rise to another deal with all other digital publishers. This is something great for the publishers but not so fascinating for Microsoft who would be attaining so much data free of cost.

Even tech giant OpenAI is on the right side of the story for Reddit, striking a deal at the year’s start for its SearchGPT.

This is all making us think about one thing. Investors can soon become tired of Microsoft’s antics and behavior because even after investing so much in AI, there’s not a lot of return. Do you agree?

We will keep you updated on whether or not Microsoft caves into the demands made by Reddit as it’s a game of wait and watch.

Image: DIW-Aigen

Read next:

• Gartner: Many Businesses Are Going to Abandon their AI Projects Because of too Much Data and Financial Risks

• New Research Shows Tech Community has Extremely Polarized Thoughts About AI and Its Implications
by Dr. Hura Anwar via Digital Information World

Gartner: Many Businesses Are Going to Abandon their AI Projects Because of too Much Data and Financial Risks

According to Gartner, a research and consulting firm, 30% of the companies who have started AI projects will abandon them by the end of next year. This is because working with AI is harder for many businesses and they lack data quality and do not have enough budgets to work with AI. Some of these businesses also do not have any idea how to manage risk controls and some of them do not have enough business value to adapt to AI.

Right now, many businesses are using AI for a number of tasks. Some are using it for coding assistance, some are taking help from AI for content creation and many are utilizing AI for document research. Some businesses are also using AI as virtual assistants while some are using LLMs for financial services. Among all these AI tasks, AI coding assistants require the least amount of cost which ranges from $100k to $200k. AI costs a fortune if you are building it from scratch for your business. If you are creating a medical or financial services LLM, it can cost from $8 million to $20 million with a recurring cost of minimum $11k to maximum $21k.

Gartner says that there was too much hype around generative AI so many companies started implementing AI last year. But most of them have realized that AI costs them a lot and they do not have enough budget to organize and develop LLMs. There are just too many risks about implementing AI in your business as compared to not implementing it at all. So, there is no wonder why many businesses are totally going to abandon this idea by the end of 2025.

Read next: Dozens Of Popular eCommerce Websites Have Defective Privacy Controls
by Arooj Ahmed via Digital Information World

Meta Reaches Massive $1.4 Billion Settlement With Texas Over Facebook’s Tag Suggestions For Photos

While the dispute may have arisen regarding a feature from more than a decade ago, tech giant Meta has finally come to a payment settlement with the state of Texas regarding its Facebook tag suggestions for pictures.

The company confirmed how it would be settling the lawsuit launched in 2022 that had to do with using facial recognition through the app. Texas stated how the feature called Tag Suggestions on images published to the app went against the state’s CUBI Act as well as the Deceptive Trade Practices law.

For this reason, Meta is all set to pay a whopping $1.4B over five years to bring the chapter to a close. This was confirmed by the Texas AG Ken Paxton today.

Certainly, it’s one of the biggest settlements related to actions brought about by a certain state as per Paxton. Furthermore, it’s also the first legal case and settlement rolled out under the CUBI which also serves as an alert to other firms that violate the Texas citizens’ privacy rights.

Any abuse of the sensitive data from Texans would be met through full investigation and force of the state law, the AG confirmed in the latest press release concerning the matter.

The problem has to do with the app making suggestions for who should be tagged in these images. This all stems from the decision that Facebook made in 2011 when it launched the latest feature dubbed Tag Suggestions. It felt this would better user experiences by ensuring it’s simpler to tag pictures with names of certain individuals in a picture as per the press release.

This was two years after we saw the state pass the CUBI Act that prevents firms from capturing biometric information such as face geometry without getting consent from them beforehand.

The app went about turning this feature for tag suggestions on without permission and that meant it was clicking and using their personal biometrics that belonged to millions of citizens without any form of authorization.

At the start, this legal case sought $25K in civil penalties for violating the law and then an additional $10K per violation of the state’s Deceptive Trade Practices Law. such penalties would have hit the hundreds of billions mark as confirmed through a report by the WSJ in 2022.

Right now, the firm didn’t admit that it was doing anything wrong in this settlement.

We did see in the year 2017 how the app went about turning this feature on/off through control tag suggestions. It was replaced with some more comprehensive settings for facial recognition, two years later. After that, the app just stopped with the automated option for face tagging images in 2021.

The spokesperson for Meta released a statement on this front, adding how they are glad that a resolution on this matter has finally been made. Moreover, he added that the company hopes to explore better opportunities in the future to better its business projects in the state of Texas like more data centers that would benefit all.

Image: DIW-Aigen

Read next: Q2 Earnings: Pinterest Revenue Matches Forecasts, MAUs Hit 522M, Shares Plunge 15%
by Dr. Hura Anwar via Digital Information World

Q2 Earnings: Pinterest Revenue Matches Forecasts, MAUs Hit 522M, Shares Plunge 15%

Social media giant Pinterest just rolled out its earnings report for the second quarter of 2024.

The news resulted in its stocks taking a hard hit of 15% as experts from the company provided insights on the upcoming quarter which was lower than what estimates from analysts mentioned.

As per reports from LSEG, the company’s performance for revenue and earnings per share included $854M and 29 cents on adjustment which were both close to the expected predictions.

Meanwhile, revenue grew 21% YoY while the net income for each quarter was $9M as compared to the loss of $35M which was equivalent to 5 cents a share as displayed in the 2023 record. During that time, the net costs and expenses hit $781M.

As reported by Pinterest, it hoped the sales for the third quarter would range anywhere between $885M to $900 million which is the equivalent to a YoY growth ranging between 16 to 18%. Experts did predict the revenue guidance to hit $907 million.

The company’s CEO was very optimistic with the earnings report, adding how the platform’s efforts for monetization are starting to pay off. Meanwhile, advertisers have seen a great improvement in performance throughout key areas of the app. This entails brand awareness to the likes of conversion.

The news comes as it keeps on launching more offerings related to AI and hence attaining a bigger share of advertising budgets seen across some of the biggest brands around the globe.

Pinterest mentioned how it included close to 522M MAU for Q2 this year which went beyond the estimates made by analysts which was 520M. As per the regional breakdown for Pinterest, the firm mentioned how the leaders in terms of userbase were certainly Europe with 136M and the American/Canadian were 98M.

Such analysts hoped the US would have a bigger share of users with predictions hitting 136M while that seen for Europe was estimated to be 284M around the globe.

In addition to that, Pinterest spoke about Q2 revenue per user to be $1.6 which again was in line with the predictions made by analysts.

Another earnings call with the company’s CFO entailed a discussion surrounding the major areas of strength for the firm and that included auto, finance, and tech to be the main stakeholders to generate revenue via the app’s online ad business. However, it was a little offset when it had to do with food and drinks as some analysts did make stronger predictions in this category.

When you look at the bigger market picture, Pinterest explained to analysts how they did see an advertising market which it felt was quite stable when compared to the results produced in the previous quarter.

Alphabet rolled out its Q2 revenue stats that went above analysts’ predictions, hitting the $84B mark while the firm’s ad sales hit $8.6B during that same time. The latter was lower than the $8.9B that estimates predicted.

Image: DIW-Aigen

Read next: Microsoft Triumphs In Q4 Again With $64.7B In Revenue Thanks To Server Products And Cloud Services
by Dr. Hura Anwar via Digital Information World

Microsoft Triumphs In Q4 Again With $64.7B In Revenue Thanks To Server Products And Cloud Services

Software giant Microsoft just rolled out its earnings for Q4 of 2024.

The financial results showed great performances related to the company’s server products and cloud services, which helped in winding up a total revenue of $64.7 billion. This included a net income of $22B during the final quarter.

Revenue growth was 15% from the previous records while net income growth was also up by 10%.

The company’s long list of server offerings and cloud services certainly get credit for this great fiscal performance as they helped generate $28.5B in total, which is a massive 19% growth seen in 2023.

Today, revenue from this particular sector makes up a whopping 45% of the overall revenue as Cloud’s earnings turn strong again.

Meanwhile, the consumer devices for Microsoft’s consumer offerings are still in decline for a repetitive seven quarters.

Coming to Windows revenue, the price paid by manufacturers for licensing Windows for different devices went up 4% YoY. As mentioned during the month’s start, the shipments for PCs grew for a repetitive three quarters which was reflected through Microsoft’s Windows OEM during the similar period.

Microsoft rolled out the latest Copilot Plus computers at the end of Q4 with an array of other devices that were powered using Qualcomm and mentioned a launch starting June 18th onwards.

On the other hand, two of the latest Surface devices were also rolled out last month. Now the change in sales that this would make wouldn’t be experienced until the start of the upcoming quarter.

Hence, this means the revenue generated through Windows and Surface went down by 11%. The final time it went up was during the first quarter of 2023.

Software giant Microsoft mentioned how the revenue for its devices would continue to fall during the upcoming quarter, hitting the lower digits as that was a true reflection of the next quarter.

Coming down to Xbox and gaming, revenue grew 61% and a lot of the credit does go to Activision Blizzard who contributed a massive 58 points to the overall scoreboard.

The company’s Xbox Game Pass offering entails 34M subscribers. The company hopes to roll out its own standard plan for this offering soon.

In other news, it also rolled out Call of Duty: Modern Warfare III through its Xbox Game Pass which would again be another attractive feature for the masses to enjoy.

But the price hikes it plans for Xbox Game Pass Ultimate will begin in September of this year with a $3 rise to hit $19.99 per month. Still, there might be a great amount of anticipation for the latest games but sales for consoles continue to be in decline. The revenue generated for hardware fell 42% in this particular quarter.

We’ve seen the software giant focusing on messages like ‘no Xbox required’ so that it could boost streaming efforts through its Cloud. Nevertheless, the revenue generated through gaming is still up 44% as a whole, thanks to the great addition in revenue by Activision Blizzard.

The revenue generated from Cloud and Office grew during the final quarter of 2024 as predicted. Meanwhile, cloud services and commercial products also rose by 12% with the revenue of Office 365 growing 13%.

Looking at the Office side of things, the revenue grew 3% YoY with subscribers hitting a 10% growth to reach 82.5M at the moment. The revenue for LinkedIn also grew by 10% in Q4.

Moving on to the company’s cloud services and server goods, the revenue generated rose by 21% while Azure and other service offerings grew by 29%. However, experts did predict the growth in this sector to be more in Q4, causing a fall in shares at NASDAQ during the after-hour trading hours.

Furthermore, we’ve got reports about investors keen on seeing any revenue growth in the AI sector of the firm. During Q4, 8 points linked to Azure and other Cloud offerings had to do with the AI services which has increased slightly than what was seen in the last quarter.

Image: DIW-Aigen

Read next:

• Google Displays The X Carousel In Search Less Frequently for News-Related Content

• CapCut Gains Massive Popularity As Video Editing App Threatens To Lure Users Away From Adobe And Canva

• The Hidden Tricks to Skyrocket Your YouTube Shorts Views!
by Dr. Hura Anwar via Digital Information World

The Hidden Tricks to Skyrocket Your YouTube Shorts Views!

YouTube Shorts are getting popular because of their quick content consumption type. You must have observed that YouTube Shorts tend to get more viral than other media because of their engaging and short nature. But this isn't all. There are some tricks YouTubers use to make their shorts get more viral and popular. Adobe Express looked at 1,000 YouTube Shorts from 100 US based YouTubers and analyzed the engagement rates, per hour views, length and tags of the Shorts. The results showed how you can optimize your Shorts to get more eyeballs.

The YouTube Shorts researchers analyzed is that all those Shorts got 218.5 million views on average which is about 12,900 views per hour. The average likes on most popular YouTube shorts are 5.4 million. The researchers found that YouTube Shorts which were longer than 41 seconds got 31% more views than shorter ones. YouTube Shorts which were about entertainment got the most views, around half (49%) of analyzed videos fall into the entertainment category.

Most YouTubers do not think about adding tags (that is all the possible labels and categories their video may fall into) on their Shorts, but after looking at 100 Shorts, it was found that most popular YouTube shorts used 24 tags on average. Using appropriate tags is important as Shorts which used relevant tags got 7% more engagement than the Shorts which used only a few tags with most of them being irrelevant. The Shorts which didn't have detailed tags also got low visibility and engagements.

Descriptions also count in getting more engagement. The most popular Youtube shorts used 60 words on average for descriptions and got 62% more comments than Shorts which didn't have any descriptions. Moreover, 94% of the most viewed YouTube Shorts used cards, 93% used high resolution thumbnails and 41% of those Shorts were also shared on Facebook. Shorts which were shared on Facebook got 29% higher rates than Shorts which weren't shared on Facebook. If YouTubers want their Shorts to get viral, it is best that they post them on other social media platforms too.

These were all the features viral YouTube Shorts had. Cross-posting on other social media apps can help a lot in bringing more engagement to Shorts, especially by sharing Shorts on Facebook. Make sure that your YouTube shorts are unique and stand out among other Shorts. Your audience matters too and it is best to create Shorts that align with your audience’s interests.

Read next: Research Shows that There's Been an Impressive Growth Seen in MarTech Products in the Last 13 Years
by Arooj Ahmed via Digital Information World