What are Deepfakes? Know how to spot manipulated videos and audio created via AI tools | How-to

What are Deepfakes? Know how to spot manipulated videos and audio created via AI tools

Deepfakes uses computer software to blend images, videos, or audio of an individual through use of AI-powered technology.

By: MEDHA JHA
| Updated on: Dec 23 2023, 07:19 IST
Google Cloud 2023 innovations: AI-powered collaborative workspaces - what it unveiled
Deepfakes
1/7 Generative AI in Google Workspace: Google Cloud introduced AI-powered features in Google Docs and Gmail, aligning with Google's AI Principles. Users can collaborate seamlessly with AI assistance in content creation, brainstorming, data analysis, and auto-generation of multimedia content.  (unsplash)
image caption
2/7 Duet AI in Google Workspace: Duet AI enhances productivity in Google Workspace, providing features like content refinement in Gmail and Google Docs, image creation in Google Slides, data insights in Google Sheets, and fostering meaningful connections in Google Meet. (pexel)
image caption
3/7 Duet AI in Google Cloud: Launched during Google I/O 2023, Duet AI in Google Cloud offers personalized experiences, including real-time code recommendations, chat assistance, and AI-powered prompts for building intelligent business applications on AppSheet.  (unsplash)
image caption
4/7 Vertex AI Search and Conversation: It empowers developers with limited ML skills to leverage Google's foundation models and search expertise. This tool enables the creation of enterprise-grade generative AI applications by combining various enterprise systems.
image caption
5/7 PaLM API, MakerSuite, and Vertex AI: Google Cloud introduced the PaLM API for safe experimentation with LLMs, MakerSuite for easy prototyping and model tuning, and new capabilities in Vertex AI, including gen AI support, three new foundation models, Embedding APIs, RLHF, and a Generative AI studio.  (unsplash)
image caption
6/7 Digital Watermarking on Vertex AI: Leveraging Google DeepMind SynthID, Google Cloud introduced advanced digital watermarking on Vertex AI, embedding watermarks directly into image pixels to make them invisible to the human eye and resistant to tampering. (unsplash)
image caption
7/7 Strategic Partnerships and Customer Collaborations: Collaborated with ONDC for a nationwide hackathon, democratizing access to digital commerce powered by gen AI; Partnered with MeitY to offer cybersecurity scholarships and train government officials;  Extended the partnership with Apollo Hospitals for the omnichannel digital healthcare experience, Apollo 24|7; Established a "Chair Professorship" in gen AI at IIT Bombay. (unaplash)
Deepfakes
icon View all Images
To spot deepfakes, there are certain anomalies left over from the video creation process that you should look out for. (Pixabay)

Deepfakes are yet again making headlines after the infamous incidents of Bollywood celebrities including Rashmika Mandana, Alia Bhatt, and others. If you've ever come across seemingly authentic but manipulated photos or videos, you may have wondered about the intriguing world of deepfakes.

How Deepfakes created?

Deepfakes are created through the use of technology to seamlessly blend existing images, videos, or audio of an individual using AI-power tools and superimposing the same on the body of another individual. At first glance, it would appear that the manipulated video is genuine as happened with the case of the Bollywood celebs mentioned above. However, these clips are totally fake and are used to defame the targeted victim.

How is it done?

This process involves training the AI software by feeding it diverse media clips of the target person, enabling it to grasp unique features akin to training facial recognition software. Once trained, the deepfake technology applies this knowledge to create entirely new, yet deceptively realistic, pictures, videos, and audio recordings.

Also read
Looking for a smartphone? To check mobile finder click here.

We are now on WhatsApp. Click to join

How to spot Deepfakes?

To spot deep fakes, follow the steps below:

1. Trust Your Instincts: Deepfakes often appear too good to be true. If a video or audio seems unbelievable, be wary.

2. Source Scrutiny: Be cautious with content from unknown or untrustworthy sources. The reliability of the source can be a key indicator.

3. Detect Inconsistencies: Observe for discrepancies in lip movements and voice synchronization. A misalignment may expose the presence of a deepfake.

4. Reverse Image Search: Verify the authenticity of the content by checking if it has been posted elsewhere online. Deepfakes are more likely to be exposed through this method.

5. Beware of Suspicious Links: Avoid clicking on links in emails or messages, especially from familiar sources, as these may be conduits for spreading deepfake content.

6. Scam Alert: Recognize signs of scams, such as requests for money or personal information, even if they are presented within a deepfake context.

7. Common Sense Prevails: Trust your instincts and apply common sense. If something feels wrong, it probably is fake.

8. Knowledge Empowers: Educate yourself about deepfakes. The more you know, the better equipped you are to spot and protect yourself from them.

What you should do about?

* Fact-Check Before Sharing: Verify information before sharing it online. Deepfake content can easily spread if not scrutinized causing much harm to the victim and it can also lead you into trouble.

* Report to Authorities: If you come across a deepfake used for malicious purposes, promptly report it to the authorities as soon as possible.

As deepfake technology advances, staying vigilant and adopting these safety measures becomes paramount in safeguarding personal privacy.

Catch all the Latest Tech News, Mobile News, Laptop News, Gaming news, Wearables News , How To News, also keep up with us on Whatsapp channel,Twitter, Facebook, Google News, and Instagram. For our latest videos, subscribe to our YouTube channel.

First Published Date: 23 Dec, 07:18 IST
NEXT ARTICLE BEGINS