Nightshade, a new tool developed by a computer science lab at the University of Chicago, may shift the power dynamics between artists and technology companies. By applying Nightshade to their work, artists can trick machine-learning models into malfunctioning by introducing “poisoned pixels.” This tool could help artists protect their work from being scraped by tech companies for training AI models without consent.
The Power of Nightshade: Empowering Artists in the Age of AI
Many artists are growing increasingly frustrated with the indiscriminate use of their visual work by AI technology companies. They are protesting against the practice of scraping their images from the internet without consent to train AI models. This power imbalance between tech companies and artists has led to a lack of control and compensation for artists.
However, a new tool called Nightshade, developed by computer science professor Ben Zhao’s lab at the University of Chicago, aims to change this dynamic. Nightshade subtly alters the pixels of an image in a way that is undetectable to the human eye but confuses machine-learning models. When artists use Nightshade on their work, the modified images become part of the AI model’s dataset and cause it to malfunction. This results in significant changes, such as dogs turning into cats or cars becoming cows. The effectiveness of Nightshade is impressive, and there is currently no known defense against it.
While some companies offer opt-out mechanisms for artists, there is no guarantee that these companies will honor their promises. Nightshade could serve as a mechanism to hold companies accountable. Building and training generative AI models is expensive, and scraping data that could potentially break these models could pose a significant risk for tech companies.
Artists like Autumn Beverly, who have experienced their work being scraped without consent, are calling for a shift towards obtaining consent and compensating artists for their contributions. They believe that this change is necessary to address the broken system of AI misuse.
Tools like Nightshade have the potential to restore power balance and protect individuals whose content is freely available online. It is a step towards ensuring that personal data, social media posts, and creative works are not misused by AI models.
Deeper Learning
How Meta and AI Companies Recruited Striking Actors to Train AI:
During Hollywood’s historic strikes, a company called Realeyes conducted an emotion study using out-of-work actors. The data collected was used to train virtual avatars for Meta, contributing to the development of more human-like AI. However, actors are concerned about the potential replacement of their profession by AI models.
Bits and Bytes
How China Plans to Judge Generative AI Safety:
The Chinese government has proposed detailed rules to determine if a generative AI model is problematic, addressing the need for AI regulation.
AI Chatbots Can Guess Your Personal Information:
Research shows that large language models can accurately predict private information from chats, raising concerns about privacy and targeted advertisements.
OpenAI Claims 99% Accuracy in Detecting Images by DALL-E:
OpenAI is developing a tool to detect AI-generated content, following a voluntary pledge by leading AI companies to develop detection mechanisms. Google has already introduced its watermarking tool.
Transparency Challenges in AI Models:
Large language models lack transparency, with the top-scoring model achieving only 54 out of 100 in Stanford University’s transparency test. Transparency is crucial for accountability and societal impact of AI.
A College Student’s AI System Deciphers Ancient Roman Scrolls:
A computer science major used AI to decipher ancient Roman scrolls damaged by a volcanic eruption. This showcases the potential of AI in historical preservation and translation.
Evolve Your Company with AI
If you want to stay competitive and leverage AI to transform your company, consider the following practical solutions:
Identify Automation Opportunities:
Locate key customer interaction points that can benefit from AI and automation.
Define KPIs:
Ensure that your AI initiatives have measurable impacts on business outcomes.
Select an AI Solution:
Choose AI tools that align with your needs and provide customization options.
Implement Gradually:
Start with a pilot project, gather data, and expand the use of AI strategically.
For AI KPI management advice, connect with us at hello@itinai.com. Stay updated on leveraging AI by following our Telegram channel t.me/itinainews or Twitter @itinaicom.
Spotlight on a Practical AI Solution: AI Sales Bot
Discover the AI Sales Bot from itinai.com/aisalesbot. This tool automates customer engagement 24/7 and manages interactions at every stage of the customer journey. Explore how AI can redefine your sales processes and customer engagement.
