🥸 South Korea declares war on deepfakes

Reading time: 5 minutes  

 Newsletter | News | Tools 

Today we will discuss-

• 🚫South Korea bans deepfake videos

• ⚖️U.S. lawmakers introduce new bill to tackle AI porn

• 😰Gen AI trend spurs privacy fears among Australians

• 😎9 amazing AI tools you might not have heard of

All this and more - Let's dive in!

👩‍🍳What’s cooking in the newsroom

South Korea bans political deepfake videos for election season

🏛️News - On 30 January, South Korea's special parliamentary committee approved an amendment to the Public Official Election Act, banning the use of AI-generated deepfake videos in political campaigns during the election season.

The move, in line with global trends, was prompted by rising concerns about the possibility of misleading content causing confusion among voters on a large scale.

🛡️Covering all bases - Under the updated law, individuals who show or share political campaign videos created with deepfake technology within 90 days before an election may face a maximum sentence of seven years in prison or a fine of nearly 50 million won ($37,618).

The revised law also requires creators to inform viewers about fake or synthetic information in deepfake videos, even if these videos are posted before the 90-day window.

🗺️What's more? Across the globe, the US Federal Election Commission has also begun the process of regulating campaign videos in preparation for the 2024 elections.

Moreover, as reported by OpenTools earlier, major tech firms like Google and Meta have updated their policies, requiring users  to include disclaimers in political ads generated using deepfake technology.

U.S. lawmakers propose bill targeting AI porn following Taylor Swift mishap

📜News - On 30 January, U.S. lawmakers proposed a bill that would criminalize  the distribution of manipulated explicit images of themselves. This move comes after pornographic AI-generated photos featuring Taylor Swift surfaced last week. 

🕵🏻‍♂️How will this help? The Disrupt Explicit Forged Images and Non-Consensual Edits (DEFIANCE) Act aims to establish a civil right of action against the creation and possession of non-consensual digital forgeries depicting identifiable individuals. 

In simple terms, this would allow victims to take legal action and collect financial damages from individuals who knowingly created or possessed such images with the intention of spreading them.

⭐What's the most important point here? The rise of sexualized deepfake images, generated by AI tools, has increased since the term was coined in 2017. Over time, off-the-shelf generative AI tools have made it even easier to produce such images, even on systems designed to prevent explicit content or impersonation. 

Over the last year, several bills related to AI and nonconsensual pornography have been proposed by Members of Congress, but most are still pending approval. More recently, the No AI FRAUD Act was proposed, aiming to introduce a broad ban on using technology to mimic someone without their consent. However, the bill raised concerns about limiting artistic expression. In comparison, the DEFIANCE Act, while still facing challenges for approval, is more limited in scope and may not pose the same risks to creative works.

Generative AI trend raises privacy concerns in Australia, Cisco study shows

👥News - The 2024 Data Privacy Benchmark by Cisco (a technology company) has revealed increased privacy concerns surrounding Generative AI (GenAI) and trust issues within businesses regarding AI. 

The study surveyed over 2,600 security and privacy professionals in 12 countries, with notable outcomes specific to Australia: 

• As per the survey, around 27% of organizations have banned the use of GenAI applications because of privacy concerns.

• 77% of respondents were concerned that the entered information might be shared publicly or with competitors.

• 69% of respondents expressed concerns about the potential harm GenAI could cause to organizations' legal and intellectual property rights.

• Additionally, 68% of organizations were concerned that GenAI might provide incorrect information to users. 

🤓In conclusion - These findings highlight the rising importance of effective governance measures to ensure data privacy and security amid growing digital threats. As AI expands across sectors, the need for stricter policies to safeguard privacy has become increasingly urgent. 

👩🏼‍💻What else is happening?

• Washington state to develop guidelines for agencies using generative AI

• Generative AI helps find merger, acquisition targets, Bain & Co. study finds

• Yelp launches revamped feed with AI-powered business summaries

• Robot trained to read braille at twice the speed of humans

• AI can better retain what it learns by mimicking human sleep

 🕰️Historic Moments

• 1961: The first industrial robot known as Unimate, which was produced by the American engineer George Devol, began working on an assembly line at General Motors in New Jersey. There, it was tasked with transporting die casings and welding parts onto cars, as these tasks were considered too dangerous for human workers.

• 1965: Carl Djerassi, an Austrian-American chemist, novelist, and playwright, created one of the earliest AI programs called DENDRAL. This program had the remarkable ability to autonomously discover previously unknown forms of medications.

• 1988: Rollo Carpenter, a computer programmer, created the chatbot Jabberwacky. The chatbot was designed to engage humans in interesting, entertaining and humorous conversations.

 😵Discover mind-blowing AI tools

1. FeedBird - A tool for creating, planning, and scheduling LinkedIn content

2. Taskade - A collaborative productivity tool for teams to plan, organize, and execute projects

3. Wonderplan - An AI-powered travel planner that generates tailored trip itineraries based on your interests, preferences, and budget

4. Teachally - An AI-powered platform designed to assist teachers in creating standard-aligned lesson plans

5. Scribble Diffusion - A diffusion tool to transform sketches into refined images

6. Tales Factory - A tool that creates custom story books with illustrations and narration

7. Snappy - Helps users select personalized gift choices for any budget or occasion

8. Airgram - A tool that helps users record, transcribe, summarize, and share meeting conversations

9. Remy Security - An AI-powered security design review tool

🎰AI-based games

• Middle-Earth: Shadow of Mordor - It is a third-person perspective action-adventure game, where you play as Talion, a ranger who seeks revenge. The game's AI-driven nemesis system creates unique interactions and relationships between the player character, Talion, and the enemies in the game based on the player's actions.

• BioShock Infinite - A first-person shooter game, set in the early 20th century in the floating city of 'Columbia'. It follows the story of Booker DeWitt, a former agent who is sent to Columbia to rescue a woman named Elizabeth. AI controls Elizabeth's actions, enabling her to contribute to the narrative and gameplay dynamics, making her an integral part of the player's journey.

• Thief: The Dark Project - A stealth game, where players assume the role of master thief Garrett. The game employs AI actors that react accurately to lights and sounds, forming a key aspect of the game's design.