AI Apocalypto Newsletter | Week 9
AI News This Week | AI ethics, Innovations & AI Apocalypto Updates!
Welcome to the 9th edition of the AI Apocalypto Newsletter! We explore the most recent developments and hot-button issues in the field of artificial intelligence this week, ranging from the EU's ground-breaking AI Act to Elon Musk's forecasts about AI overtaking human intellect.
AI Apocalypto hosted their 1st AI and Blockchain meetup in Hanoi. And revealed their AIverse story & NFT art!
Ethical Usage of AI
Ethical AI strives for fairness and transparency, shaping a responsible AI future.
Emad Mostaque Steps Down as Stability AI CEO
• In pursuit of decentralized AI.
• COO Shan Shan Wong and CTO Christian Laforte serve as interim co-CEOs.
• Stability AI faces lawsuits over Stable Diffusion's data.
• Most Inflection staff join Microsoft AI.
EU's AI Act Approved
• It bans certain AI uses.
• Requires transparency from AI companies.
• Establishes a complaints system.
• Aims to regulate AI's impact.
• Critics question facial recognition and algorithmic transparency.
Ethical AI for Children
• Advocates for thoughtful AI implementation.
• Highlights challenges in applying high-level principles.
• Stresses the need for child-centered evaluations.
• Advocates for fair access, transparency, privacy protection, and age-appropriate systems.
UN General Assembly Resolution on AI
• Highlights AI's benefits and drawbacks.
• Calls for global standards.
• Emphasizes positive potential, digital transformation, and equitable access.
• Aim to support the UN's Sustainable Development Goals by 2030.
• Warns against AI misuse and biases.
Unethical Usage of AI
Unethical AI raises concerns over misuse and bias, highlighting ethical challenges.
French Regulators Ruling Google for Media Algorithms
• Google was fined 250 million euros for failing to negotiate fair licensing deals.
• Critics criticized for using news articles without content prevention until September.
• Google disputes publishers over payment for news content display.
AI Models' Racism Against African American Vernacular English
• ChatGPT and Google's Gemini show covert racism.
• Perceive AAVE speakers as unintelligent, and lazy.
• Recommend harsher punishments.
• AI use raises discriminatory concerns.
AI Chatbot Security Threats
• AI assistants are vulnerable to hacking despite encryption.
• Study reveals sophisticated attack deciphering AI responses.
• Attack exploits side channels and refines results using large language models.
• Mitigation strategies include reevaluating token transmission protocols.
Lego Criticized for AI-Generated Artwork
• Lego used AI-generated artwork in the Ninjago quiz.
• Images included elements from other intellectual properties.
• Images removed.
• Lego reaffirmed its policy against AI use.
• Incident sparked debate on AI's impact on creativity and copyright.
OpenAI ChatGPT Store Overrun
• Overrun due to spam and copyright issues.
• Bots claim to generate Disney/Marvel art.
• Bots simulate public figures' conversations without consent.
• OpenAI policy prohibits academic dishonesty and impersonation.
Chatbot Security Vulnerability
• Researchers warn of 55% accuracy in inferring chat content.
• OpenAI's encryption methods criticized.
• Microsoft assures personal details are unlikely to be compromised.
AI Advancements
Advancements in AI push boundaries, transforming industries and enhancing human capabilities.
Sakana AI Releases Evolution-Inspired AI Models
• Tokyo-based startup introduces evolution-inspired models.
• Uses "model merging" to combine existing models with evolution approaches.
• Identifies successful models as "parents" for future generations.
• Releases three Japanese language models, two open-sourced.
MIT's FeatUp System Enhances Computer Vision Algorithms
• Adjusts images for high-resolution features.
• Improves object detection and depth estimation.
• Potential for widespread adoption.
• Enhances model detail perception.
Impact of GenAI on Business Value Delivery
• GenAI boosts operational and business value expectations.
• Asia Pacific companies plan a 45% increase in AI expenditure in 2024.
• CIOs lead GenAI investments.
• Gartner predicts a global AI market surge.
• AI primary in BFSI sector risk and fraud detection.
OpenAI's GPT-5 Launch
• Mid-year launch expected.
• CEO Sam Altman hinted at multiple releases.
• Release timeline uncertain.
• GPT-5 in a training phase, safety testing, internal "red teaming."
• Uncertainty over copyrighted training data handling.
VTouch Inc. Launches WIZPR Ring
• First-of-its-kind smart wearable ring with advanced AI capabilities.
• Allows discreet whispering of commands, filtering out background noise.
• Allows switch between AI assistants via button press or scenario-specific aid.
• Connected via Bluetooth for efficient conversation.
Mia AI Detects Breast Cancer Signs in Mammograms
• Piloted with NHS clinicians.
• Identified 11 missed cases.
• Trained on 6,000 cases.
• Predicted cancer presence with 81.6% accuracy.
• Aims to predict post-treatment side effects.
AI Apocalypto Updates
AI meets Blockchain Meetup in Hanoi, Vietnam
• Hosted by AI Apocalypto, Xrender, Mamoru.ai and AI MetaFi in partnership with Superteam Vietnam.
• AI Apocalypto revealed their 1st Episode of their AIverse story in the Product reveal
• Engaging in discussions on NFTs and decentralized technology.
• Latest Blockchain + AI tech were showcased.
• Provided networking opportunities and Exciting swags.
• The event took place on 19th March 2024 at MAC Plaza, Hanoi.