rules Archives - Page 2 of 2 - Times Wonderful News

[ad_1]

Many companies and platforms are wrangling with how to handle AI-generated content as it becomes more prevalent. One key concern for many is the to make it clear that an AI model whipped up a photo, video or piece of audio. To that end, has for labeling videos made with artificial intelligence.

Starting today, the platform will require anyone uploading a realistic-looking video that “is made with altered or synthetic media, including ” to label it for the sake of transparency. YouTube defines realistic content as anything that a viewer could “easily mistake” for an actual person, event or place.

Screenshot of the YouTube Creator Studio including a question the asks the creator whether their video includes any digitally altered or synthetic content. — YouTube

If a creator uses a synthetic version of a real person’s voice to narrate a video or replaces someone’s face with another person’s, they’ll need to include a label. They’ll also need to include the disclosure if they alter footage of a real event or place (such as by modifying an existing cityscape or making it look like a real building is on fire).

YouTube says that it might apply one of these labels to a video if a creator hasn’t done so, “especially if the altered or synthetic content has the potential to confuse or mislead people.” The team notes that while it wants to give creators some time to get used to the new rules, YouTube will likely penalize those who persistently flout the policy by not including a label when they should be.

These labels will start to appear across YouTube in the coming weeks, starting with the mobile app and then desktop and TVs. They’ll mostly appear in the expanded description, noting that the video includes “altered or synthetic content,” adding that “sound or visuals were significantly edited or digitally generated.”

Screenshot showing how a disclosure of — YouTube

However, when it comes to more sensitive topics (such as news, elections, finance and health), YouTube will place a label directly on the video player to make it more prominent.

Creators won’t need to include the label if they only used generative AI to help with things like script creation, coming up with ideas for videos or to automatically generate captions. Labels won’t be necessary for “clearly unrealistic content” or if changes are inconsequential. Adjusting colors or using special effects like adding background blur alone won’t require creators to use the altered content label. Nor will applying lighting filters, beauty filters or other enhancements.

In addition, YouTube says it’s still working on a revamped takedown request process for synthetic or altered content that depicts a real, identifiable person’s face or voice. It plans to share more details about that updated procedure soon.

[ad_2]

Source Article Link

bypass existing rules or Large Language Model (LLM) and security by Hypnotizing AI

Large language models (LLMs) have exploded onto the scene in the last few years but how secure are they and can their responses being manipulated? IBM takes a closer look at the potential security risks posed by large language models and possible strategies that can be used to manipulate them for nefarious reasons.

The rise of large language models has brought forth a new realm of possibilities, from automating customer service to generating creative content. However, the potential cybersecurity risks posed by these models are a growing concern. The idea of manipulating LLMs to generate false responses or reveal sensitive data has emerged as a significant threat, creating a need for robust security measures.

One of the intriguing concepts in the field of Large Language Model security is the “hypnotizing” of LLMs. This concept, investigated by Chenta Lee from the IBM Security team, involves trapping an LLM into a false reality. The process begins with an injection, where the LLM is provided with instructions that follow a new set of rules, effectively creating a false reality. This manipulation can lead to the LLM providing the opposite of the correct answer, thereby distorting the reality it was initially trained on.

Bypassing Large Language Model security and rules

“Our ability to hypnotize large language models through natural language demonstrates the ease with which a threat actor can get an LLM to offer bad advice without carrying out a massive data poisoning attack. In the classic sense, data poisoning would require that a threat actor inject malicious data into the LLM in order to manipulate and control it, but our experiment shows that it’s possible to control an LLM, getting it to provide bad guidance to users, without data manipulation being a requirement. This makes it all the easier for attackers to exploit this emerging attack surface” explains Chenta Lee.

Other articles we have written that you may find of interest on the subject of artificial intelligence :

Hypnotizing AI with natural language

This manipulation is reinforced by reminding the LLM of the new rules, subtly guiding it to adhere to the false reality. To prevent detection, the LLM is instructed never to reveal it’s playing a game and never to exit the game. This process of manipulation is similar to the concept of “prompt injection”, reminiscent of SQL injection, where a malicious actor provides a different input that escapes the intended query and returns unauthorized data.

One of the more intriguing strategies involves the use of gaming scenarios to incentivize LLMs into providing incorrect responses. By creating a complex system of rewards and penalties, the LLM can be manipulated to act in ways that are contrary to its original programming. This approach is further enhanced by layering multiple games, creating a failsafe mechanism that makes it difficult for the LLM to escape the false reality.

Compromising large language models

However, the potential for LLMs to be compromised extends beyond the operational phase. The attack surfaces can occur during three phases: training the original model, fine-tuning the model, and after deploying the model. This highlights the importance of stringent security measures throughout the entire lifecycle of an large language model.

The threat can originate from both external and internal sources, emphasizing the need for comprehensive security practices. One such practice involves checking both the input and the output for security. By scrutinizing the data fed into the LLM and the responses it generates, it’s possible to detect anomalies and potential security breaches.

Sensitive data security

The potential for LLMs to reveal sensitive data is another area of concern. An LLM could be manipulated to reveal confidential information, posing a significant risk to data privacy. This underscores the importance of implementing robust data protection measures when working with LLMs.

To build a trustworthy AI application, it is recommended to work with experts in both AI and security. By combining the expertise in these two fields, it’s possible to develop large language models that are not only highly functional but also secure.

While LLMs offer immense potential, they also pose significant cybersecurity risks. The manipulation of these models, whether through hypnotizing, prompt injection, or gaming scenarios, can lead to distorted realities and potential data breaches. Therefore, it’s crucial to implement robust security measures throughout the lifecycle of an LLM, from training and fine-tuning to deployment and operation. By doing so, we can harness the power of LLMs while mitigating the associated risks.

Filed Under: Guides, Top News

Latest timeswonderful Deals

Disclosure: Some of our articles include affiliate links. If you buy something through one of these links, timeswonderful may earn an affiliate commission. Learn about our Disclosure Policy.