August 23, 2025 | 10:30 am
TEMPO.CO, Jakarta - Deepfake is a digital manipulation in the form of photos, videos, or audio that have been modified to look extremely realistic. This technology falsely depicts someone doing or saying something they never actually said or did.
By utilizing the concept of deep learning in artificial intelligence (AI), deepfake is able to generate content with high realism, complete with voice imitations, intonation, and movements that are difficult to distinguish from the original.
The presence of this technology raises concerns about privacy, identity theft, and misinformation. The existence of deepfakes further blurs the line between fact and fiction, and they're frequently abused for various crimes, such as identity theft, fraud, biometric system breaches, and fake news.
Many people think they are safe because they don't engage with AI products, but this technology can collect data from millions of people through the internet, including photos, videos, and voice recordings circulating on social media.
Although it is difficult to avoid completely, there are steps to reduce the risk. The simplest way is to be careful about what is shared publicly.
As reported by NSW Government, here are some tips to protect yourself from the threat of deepfakes.
1. Be Wise in What You Share Online
The first step to protect yourself from the threat of deepfakes is to be cautious in sharing personal information on the internet. Limit the personal information published, especially high-quality photos and videos. Set social media privacy so that only trusted people can see the posts.
2. Strengthen Data Privacy and Security
Use privacy settings on various platforms to restrict who can access your personal photos, videos, or information. When sharing images or videos, consider adding a digital watermark so that the engineering trail can be easily traced.
3. Use Multi-Factor Authentication and Strong Passwords
Protect important accounts with multi-factor authentication (MFA). With this method, the login requires additional steps, such as facial scanning, unique codes via SMS, or specialized authentication applications. Also, use long, strong, and unique passwords of at least 16 characters, and securely store them using password managers equipped with MFA.
4. Regularly Update Devices and Beware of Phishing
Ensure that devices and applications are always updated automatically so that security loopholes are not exploited by hackers. Also, be wary of suspicious messages through email, SMS, or phone, especially if they demand an immediate response, such as claiming prizes or warning of hacking.
5. Be Skeptical
Remain skeptical of viral videos or audios on social media, especially ones with unclear sources. Immediately report deepfake content that implicates individuals or close associates.
Editor’s Choice: The Different Types of Deepfakes You Should Know About
Click here to get the latest news updates from Tempo on Google News
YouTube Rolls Out AI Age Verification to Flag Users Under 18 in the U.S.
1 hari lalu
YouTube accounts identified as belonging to minors will receive additional protections, such as the blocking of age-restricted videos.
The Different Types of Deepfakes You Should Know About
1 hari lalu
Understanding deepfakes, including the various types, is one way to combat them.
Piping Down the Hype over Artificial Intelligence
1 hari lalu
evolution of science has been central to the narratives around computers, genetic engineering and artificial intelligence (AI).
The Rise of AI-Generated Music and How to Recognize Them
2 hari lalu
AI-generated music often confuses casual listeners as many of its songs sound so realistic that they are difficult to distinguish from human compositions.
Texas Attorney General Probes Meta, Character.AI over Misleading Mental Health Claims
3 hari lalu
The Attorney General of Texas is investigating chatbot owned by Meta and Character.AI for allegedly engaging in deceptive practices by marketing themselves as mental health tools.
Japanese Consortium Develops AI for Information Verification
4 hari lalu
A consortium in Japan, including Fujitsu Ltd., is developing an AI-based fact-checking platform. What is the purpose of this project?
Here Are 10 European Money Tips Worth Remembering
5 hari lalu
Whether you're on a high-end vacation travel or a quick getaway in Europe, these tips will help you keep control of your spending.
Elon Musk's Accusations Against Apple Prompt Responses from Sam Altman, OpenAI's CEO
6 hari lalu
Allegations made by Elon Musk have been responded to by Apple and Sam Altman, the CEO of OpenAI
Komdigi Calls for Public Input on Indonesia's AI Roadmap
7 hari lalu
There are four focuses to be developed in the AI roadmap according to the Guidelines for the 2025-2029 National Medium-Term Development Plan (RPJMN).
Russia Requires Switch to MAX, Dropping WhatsApp and Telegram
8 hari lalu
Technology experts suspect that the Russian government is using the MAX app to control the internet.
