type here...
Home Blog Page 57

Doctored Kamala Harris Video Shared by Elon Musk on X

Indira Akmalia
-
0
Doctored Kamala Harris Video Shared by Elon Musk on X
Reading Time: < 1 minute read

posted a doctored Kamala Harris campaign video to X over the weekend. The video contains a deepfake voiceover speaking over it that makes misleading claims. Nowhere does Musk’s post mention that the video is altered in any way. By X’s policies, users are not allowed to share misleading media, but the video remained on the platform all weekend, viewed over 119 million times.

The video, which was originally posted by @MrReaganUSA, purports to be a parody. The voiceover makes false claims about Harris and Joe Biden. Musk’s post is simply, “This is amazing,” followed by a laughing emoji. It has not been labeled as misleading, despite X’s policy. No Community Notes have been added, though many were suggested.

Some manipulated media may remain up under X policies, such as memes and satire, provided they do not confuse users. The possible misuse of deepfakes is a concern in its nascent stage. It can influence the votes in the run-up to elections. As recently as this year, 20 tech companies signed an agreement to tackle this.

Deepfake videos are extremely harmful in the present digital era, they can easily sow confusion and deceive public opinion. The platforms are on high alert since elections are nearing. A check has been put on ensuring that the media being shared are authentic. Be careful; users might be tricked.

Apple Intelligence Delayed: Coming with iOS 18.1 in October

Indira Akmalia
-
0
Apple Intelligence Delayed: Coming with iOS 18.1 in October
Reading Time: 2 minutes read

Apple Intelligence won’t be ready for iOS 18 to drop in September. Now for the good news, the wait isn’t long either. We should expect Apple Intelligence to release iOS 18.1—which should go public in October according to the company.

Developers will soon get a taste of what’s new. Those betas include a subset of AI features and are due out this week. Overall, Apple always aims for developer experimentation with major features early on as it helps the company streamline the technology for public release.

  • Apple Intelligence
  • Apple Intelligence
  • Apple Intelligence

Some of the features, however, are not going to be immediately available. An AI-powered Siri, for instance, will come in later this year. Some of these features may slip into next year. Apple would rather they work perfectly.

One such factor is compatibility. The Apple Intelligence will require the current bearings of devices using the A17 Pro processor or the M-series chip. The recent models of the iPhone are not supported, only the iPhone 15 Pro and 15 Pro Max will support Apple Intelligence initially.

Apple’s move is very purposefully exclusive. High-powered hardware is a requisite part of the package that has the upgraded artificial intelligence features built into it. It assures that the user experience is protected. Thus, less capable, older devices may not be fully used in terms of Apple Intelligence.

Apple Intelligence with iOS 18.1 will bring breakthrough features. Though they are a tad bit late, the anticipation keeps increasing. Both developers and users curiously await a glance at new possibilities. Again, this clearly states that Apple is more focused on quality and compatibility. For sure, the waiting is going to be worth it.

The New ChatGPT Voice Mode Features for Plus Subscribers

Indira Akmalia
-
0
The New ChatGPT Voice Mode Features for Plus Subscribers
Reading Time: < 1 minute read

Beginning next week, a major update hits ChatGPT with a new Voice Mode, furthering the offerings available at OpenAI. All of the members who have subscribed to the facility of ChatGPT Plus will have these new features made available to them. This update, originally targeted for release in June, was delayed and is now back on track. It remains to be seen whether or not this update will roll out to all Plus members at once.

This update to the voice mode of ChatGPT comes with a couple of exciting features. Most importantly, this voice mode can offer users realistically paced voice conversations since its response time is low. Thus, this change makes the interaction with ChatGPT much more natural and engaging.

The second feature is live translation between languages. It allows one to communicate across languages seamlessly, furthering the versatility of ChatGPT. One of the more interesting new features is the ability to create sound effects while narrating. This feature will possibly boost user engagement with the addition of an auditory dimension for interaction with ChatGPT.

It also adds custom character voices, so now users will have the ability to hear what ChatGPT has to say in a wide variety of character voices, giving it a funny and personalized feeling when speaking. This update follows the prior announcement of SearchGPT, OpenAI’s new search engine, and provides enhanced capabilities that place ChatGPT at the top of its game for users.

YouTube Unblockable Ads: New Policy Changes Explained

Indira Akmalia
-
0
YouTube Unblockable Ads: New Policy Changes Explained
Reading Time: 2 minutes read

YouTube is cracking down on ad blockers. Recent reports suggest the platform tests server-side ad injection. The system includes putting ads directly in videos, making blocking very hard. Users say these are appearing in their video streams as black screens. This happened at the same time when Google announced stricter policies for browser extensions.

YouTube Ongoing Campaign Against Ad Blockers

Previously, YouTube issued warnings to those who used ad blockers. This past summer, the site limited access to videos until ads were turned on. In the fall, it applied a hard block with a message imploring users to turn off ad blockers or to subscribe to YouTube Premium. All these actions are part of YouTube’s continuing campaign against ad-blocking technologies.

According to reports, this month it seems like YouTube started pushing that strategy. According to a post on the subreddit thread, server-side ad injection of the site was now said to be seen in the wild. The move provides a more robust solution against ad blockers by directly embedding ads into video streams. Tests were also going on, as indicated by the black screen issue some users encountered.

The timing of all this is interesting. The new ad injection technique was detected on the very same day that Google updated its policy on browser extensions—a coincidence that strongly indicates a coordinated effort to clamp down on ad blocking. The policy change is also designed to enhance the restrictions on browser extensions, which are often used to block ads.

YouTube

As innovations in ad strategies on YouTube come in, new challenges are presented to users. Integrating ads into the video stream has created a new obstacle for ad blockers. This may now be the turning point of the strategy of YouTube regarding advertisements and user engagement.

YouTube’s actions foreshadow something huge in the approach toward ad blocking. Tighter policies against ad blockers and new techniques are making the platform fight back. Users might want to adapt to these evolving ad strategies if they want to enjoy their content uninterrupted.

New Google Gemini Beta: Fine-Tuning Capabilities for AI Images

Indira Akmalia
-
0
New Google Gemini Beta: Fine-Tuning Capabilities for AI Images
Reading Time: 2 minutes read

It is a very powerful AI-based chatbot developed by Google Gemini that generates pictures according to a given text prompt. At this point, the system cannot allow image tweaking once it has been generated. The latest version of Google’s beta app for Android suggests new fine-tuning tools for AI image generation. Users will be able to modify images without starting from square one using this feature.

Gemini

These fine-tuning tools provide two major ways to customize his images. First, one can change the details of the image by specifying natural language prompts. This technique will ensure that the core of the original image remains the same but with specified changes.

Gemini

The second is by circling the part of the image that a user wants to edit. In this way, users will be able to specify changes without explaining an image in detail. In both ways, the main purpose is to save time and facilitate users properly.

For now, these features are not fully usable. Google may be flipping them on via server-side updates. Still, this change is a rather positive omen, hinting at a future enhancement that should soon arrive in Gemini. Tweaking AI images will become more streamlined and user-friendly.

Grok Chatbot Under Fire for Spreading Election Misinformation

Indira Akmalia
-
0
Grok Chatbot Under Fire for Spreading Election Misinformation
Reading Time: < 1 minute read

Minnesota Secretary of State Steve Simon has come out against the chatbot, X’s Grok. Grok has been notorious for spreading false information about ballot deadlines related to the presidential race, according to Simon. The Minneapolis Star Tribune reports that Grok was telling people that multiple states had missed their deadlines, including Minnesota, Texas, and Pennsylvania.

It’s a falsehood, originally from a post by user @EvanAKilgore on X. Grok’s “Fun Mode” is even eminently sarcastic about it, saying candidates should time-travel. That is misleading—President Biden’s name is not locked on ballots. States don’t begin printing ballots until after party conventions.

Minnesota’s ballot deadline isn’t until August 26, not past. Fact-checkers such as Politifact have disposed of the theory that Biden’s name is already cemented. Grok has persisted with posting the flawed information, yet no corrections were made.

Simon’s staff did attempt to inform X of the misinformation, but there were some problems in the communication. They got, as they termed it, a “shoulder shrug.” The immediate return from X’s press email did not help or clarify things.

This points to a big issue, there is just no feedback from X for this right now. Information provided by AI chatbots can be devastating if it is inaccurate. This incident underlines the necessity of more control and accuracy in AI-powered tools.

New Apple Music Feature: AI-Generated Playlist Art in iOS 18

Indira Akmalia
-
0
New Apple Music Feature: AI-Generated Playlist Art in iOS 18
Reading Time: < 1 minute read

One of the new features that Apple could add to iOS 18 would be AI-generated playlist art for Apple Music. According to 9to5Mac, this functionality was found within the iOS 18 beta code. A button might say “Create Image” when one edits playlists, which would allow custom art to be created with AI. It will work almost in the same way the new iOS version does while creating AI-generated emojis.

The functionality does not exist in the iOS 18 beta, so we don’t know exactly how it works. We’re not even sure that it will make its way into iOS 18 when it’s given a full public release, Bloomberg’s Mark Gurman hinted some AI features might be pushed back to 2025.

Even so, mentioning AI-generated playlist artwork does make things very exciting for the future. Further enhancing user experience with customized artwork is more importantly indicative that iOS 18 may include many more innovative AI features.

Moreover, enhanced image and video recovery features might also make their way into iOS 18. If that happens, then such features shall truly make this update worth it. All in all, iOS 18 is offering much more than what was unveiled.

All this and more will be a possibility with subsequent updates. Apple’s firm commitment to the integration of AI into its products is becoming more and more evident.

OpenAI Launches SearchGPT: A New Era of AI-Powered Search

Indira Akmalia
-
0
OpenAI Launches SearchGPT: A New Era of AI-Powered Search
Reading Time: 2 minutes read

OpenAI recently opened access to its AI-based search engine, SearchGPT, to some users. It, of course, wants to integrate the best characteristics of SearchGPT into ChatGPT down the road. The release is part of OpenAI’s ongoing pursuit of deploying its AI tools in everyday life.

How to Try SearchGPT

Now, users can register to be able to try SearchGPT, though testing has not begun. The search engine is similar to ChatGPT, allowing users to type their queries and then return results intended to be both pleasing to the eye and results-driven—but users can then respond to results with further questions for more refined answers.

SearchGPT

Though exciting, there is concern over the accuracy of SearchGPT. Kif Leswing from CNBC pointed out that OpenAI example results included several inaccuracies. OpenAI claims to include a clear link from where the information came in the search result. How well this feature works is yet to be determined.

Promising First Step

Although OpenAI still has a long way to go before it can stand shoulder-to-shoulder with Google, SearchGPT is an important first step. Drawing the ability to follow up with conversational queries could make searches more intuitive.

SearchGPT may redefine how users view search engines. The facility’s inclusion in ChatGPT, already an extremely popular tool in its own right, would combine to provide users with a seamless experience across a variety of AI tools. In the development process, focusing on enhancing the accuracy and usability of OpenAI would be imperative.

The arrival of SearchGPT is evidence that OpenAI sincerely believes in the growth of AI technology. Though there are risks involved, the possible reward is huge. Thus, many people are very excited about being provided with the opportunity to test and see this new AI search engine.

WhatsApp Surpasses 100 Million Monthly Users in the U.S.

Indira Akmalia
-
0
WhatsApp Surpasses 100 Million Monthly Users in the U.S.
Reading Time: 2 minutes read

According to Meta CEO Mark Zuckerberg, WhatsApp has crossed 100 million monthly active users in the United States. This is the first time that data regarding its American market has been shared. Of interest, over half of WhatsApp users in the United States are iPhone owners, mirroring the market lead at this point held by iOS.

Globally, WhatsApp still holds the position with a user base of more than 2 billion monthly active users. India has the maximum number of WhatsApp users, with more than 500 million monthly active users, followed by Brazil and Indonesia. Its U.S. growth, however, is indicative of its growing influence.

WhatsApp

WhatsApp has recently increased its marketing efforts across the U.S. The company has run several ad campaigns where it promotes the features on Times Square billboards and TV. One campaign running recently featured cast members of the TV show “Modern Family” in an attempt to increase user engagement.

In its marketing strategy, Meta has very often pitted WhatsApp against the likes of Apple’s iMessage. However, despite this difference between the green bubble and the blue bubble, WhatsApp is never behind. Now, Apple is adopting RCS, providing Android users with advanced media capabilities and read receipts, but messages remain green.

WhatsApp

In addition, it has higher-order features such as polls, high-quality video calls, and reactions. Meta is closing out its marketing drive by placing the largest bubble ad—200 feet between Apple and Samsung stores—at The Americana Mall in Los Angeles as a catching visual promotion for user adoption.

How this has grown its U.S. user base is a combination of good marketing and feature improvements. Meta, to date, sustains WhatsApp at the top among messaging apps through conventional and innovative promotional schemes.

Gemini AI Expands Globally with New Features

Indira Akmalia
-
0
Gemini AI Expands Globally with New Features
Reading Time: < 1 minute read

A big update has hit Google’s Gemini AI chatbot in the form of Gemini 1.5 Flash which speeds up response times and enhances reasoning. Image understanding also improves, and a quadrupled context window is available.

Gemini AI

Previously, only Gemini Advanced allowed the use of the 32K token context window, now it’s free for all. Long conversations and complex questions are now possible. Soon enough, Gemini will be able to upload files through Google Drive or from your device.

Another important addition is that of “related content.” The links related to relevant websites or emails will now show up in Gemini’s responses, reducing hallucinations and encouraging further research on the subjects.

It also increases its geographic reach. It is available in more than 230 countries and supports 40 languages. Gemini AI will also be integrated into Google Messages on select Android devices in the EEA, UK, and Switzerland.

It will give the world’s teenagers access to Gemini in over 40 languages very soon. This aims to help with school subjects and university preparation. Google puts more policies for safe and responsible usage into practice.

It says that during the development of Gemini, Google has truly kept its responsibility to users at the core of its development. They have now published more details regarding how Gemini has been designed and on response guidelines. According to Google, they will be sharing more news regarding Gemini during their Made by Google event next month.

1...565758...212Page 57 of 212
© Newspaper WordPress Theme by TagDiv