- AI Odyssey
- Posts
- Predictions for AI in 2024
Predictions for AI in 2024
PLUS: AI cannot patent inventions, UK Supreme Court confirms.
Hello! Festive heads-up! ๐ This newsletter is our last blast until January. Catch you on the flip side of the holidays! ๐โจ
โฐ Top 4 Headlines
๐Link: AI cannot patent inventions, UK Supreme Court confirms.
๐Link: New AI can predict peopleโs time of death with high degree of accuracy, study finds.
๐Link: OpenAI buffs safety team and gives board veto power on risky AI.
๐Link: ChatGPT now lets you archive chats. The feature is out now on the web and iOS, and OpenAI says itโs coming to Android โsoon.โ
8 predictions for AI in 2024
As we step into 2024, the AI realm is poised for a transformative journey. Here's a rundown, sans the jargon, of what to expect:
OpenAI's commercial pivot: OpenAI is undergoing a facelift, steering towards a product-centric approach, with the GPT store emerging as a focal point.
AI's creative evolution: Watch out for advancements in agent-based models and generative multimedia, offering practical solutions that extend beyond the experimental stage.
Reality check for language models: Large language models face scrutiny, prompting a quest for more efficient alternatives as their limitations become increasingly evident.
AI marketing woes: The hype surrounding AI tools may hit a snag as exaggerated claims lead to customer discontent, potentially triggering legal and regulatory responses.
Apple's entrance into AI: Apple, the perennial observer, is poised to enter the AI arena, focusing on practical applications and delivering standout agent capabilities.
Legal wrangles over AI: Expect a surge in lawsuits related to AI misuse and bias, with companies opting for settlements to sidestep divulging sensitive information.
AI embracing regulations: Companies are gearing up for compliance with emerging AI regulations, setting the stage for the growth of an AI compliance industry.
AI's impact on the 2024 election chaos: Brace yourself for an AI-fueled tumult in the 2024 U.S. presidential election, where generated content and analyses play a pivotal role in shaping public opinion and election dynamics.
Rite Aid hit with five-year facial recognition ban over โrecklessโ use
Rite Aid faces a five-year ban on using facial recognition tech, following a settlement with the Federal Trade Commission (FTC).
The FTC accuses Rite Aid of reckless use of facial surveillance systems from 2012 to 2020, creating a database of customers flagged for suspicious behavior. The complaint reveals Rite Aid's failure to inform customers about facial recognition use, disproportionately affecting people of color. The pharmacy chain must now implement safeguards, delete all collected photos and videos, and provide notice for future biometric data enrollment. The FTC's order awaits approval as Rite Aid undergoes bankruptcy proceedings. Other retail stores also face scrutiny for facial recognition use.
Researchers warn AI could add billions to U.S. racial wealth gap
Generative AI could amplify the U.S. racial wealth gap by $43 billion annually over the next two decades, with Black workers facing heightened job loss risks due to automation.
McKinsey researchers warn that generative AI might contribute an alarming $43 billion yearly to the existing racial wealth gap, underscoring the widening economic divide.
Job threat from AI: The report highlights AI's threat to jobs requiring over $42,000 annually and not demanding a 4-year degree. Professions at risk, like office support and production, disproportionately impact Black workers.
Persistent inequalities: Existing disparities, from lower pay for Black Americans to limited tech access and underrepresentation in AI development, compound the challenges. Historical biases, including redlining, further hinder opportunities.
As AI evolves, proactive measures are crucial to prevent deepening inequalities. Recommendations include upskilling for Black workers, employer-backed reskilling, and AI developers prioritizing bias-free models. The report sees potential for positive impact with proper governance, fostering economic mobility and improved healthcare access.
Stability AI unveils paid memberships for commercial AI access
Stability AI, known for its text-to-image model Stable Diffusion, is rolling out a paid membership for commercial AI use.
Membership tiers: Stability AI offers a free tier for personal use, a $20/month tier for creators and startups, and an enterprise plan for commercial use. Only paid tiers get early access to new AI models.
Profit-openness balance: The company aims to redefine how it grants commercial rights, seeking a balance between profitability and openness, echoing its commitment to transparency.
Pricing puzzles: Users are puzzled about commercial use details and potential fees. CEO Emad Mostaque admits they're simplifying pricing while users on the subreddit express confusion over AI-generated work fees.
Stability AI's move into paid memberships marks a commitment to advancing AI research.