Dimensionless Tech - Tech News Updates #10

OpenAI Unveils GPT Store: A Marketplace for Custom AI Assistants

• OpenAI has launched the GPT Store, enabling users to access and share custom AI assistants created using GPT technology. 

• Within two months, users have generated over three million custom assistants, highlighting the rapid adoption of this technology.

• The store features diverse assistants in art, research, programming, education, lifestyle, etc. Notable ones include AllTrails, Consensus, Khan Academy’s Code Tutor, Canva, and CK-12 Flexi.

• OpenAI is enabling users to create assistants without coding skills. Starting Q1 2024, eligible US-based builders can earn money by monetizing user engagement with their assistants through payment incentives.

Rabbit Inc. Unveils R1, CES 2024

• Rabbit Inc. debuts the R1 at CES 2024, a pocket-friendly device designed to simplify daily tasks without the need for smartphone apps.

• R1 introduces a push-to-talk button for seamless task automation, eliminating the need for app navigation.

• Operating without onboard apps or API connections, Rabbit OS streamlines tasks through a web portal, ensuring a clean user experience.

• R1 prioritizes user privacy with no stored credentials, direct authentication on third-party services, and a push-to-talk feature that activates the microphone only when needed.

• A standout feature, "teach mode," enables users to train their own automated scripts, adding a personalized touch to task automation.

Deep Fake Dilemma: Over 100 AI-Generated Advertisements Target UK Prime Minister on Facebook

• Fenimore Harper's research reveals over 100 deep fake video ads featuring UK Prime Minister Rishi Sunak promoted on Facebook, reaching around 400,000 people through 143 adverts.

• Deep fake origins traced to countries like the US, Turkey, Malaysia, and the Philippines. Ad expenditures surpassed £12,929, exposing the international scope and financial investment in such deceptive campaigns.
• Fenimore Harper founder highlights the ease of using AI for malicious purposes, emphasizing insufficient moderation of paid advertising on platforms like Facebook. Many of the ads violating policies remain active.
• The UK government acknowledges the issue and expresses commitment to defending democratic processes. The Online Safety Act aims to impose stricter requirements on social platforms to swiftly remove AI-generated misinformation.

OpenAI's Copyright Conundrum: AI Advancement vs. Legal Challenges

• OpenAI asserts that developing advanced AI systems, such as ChatGPT, would be "impossible" without utilizing vast amounts of copyrighted data, citing the impracticality of adhering to expansive copyright laws.
• While OpenAI suggests potential partnerships and compensation schemes with publishers to support creators, there is no indication of a significant shift in its data collection practices, including access to paywalled content.
• OpenAI's stance has triggered lawsuits, including from media outlets like The New York Times, alleging copyright breaches. The company faces legal battles around AI systems designed to absorb large volumes of protected text and media.

DeepMind Unleashes Advanced Tools for Autonomous Robot Learning

• DeepMind unveils AutoRT, using Large Language and Visual Models for robots to autonomously assess scenes, generate tasks, and prioritize based on safety and diversity.

• DeepMind orchestrates 20 robots, conducting 77,000 trials on 6,650 tasks in diverse offices, highlighting AutoRT framework's effectiveness.

• Self-Adaptive Robust Attention for Robotics Transformers (SARA-RT) improves efficiency by 14% and accuracy by 10%, addressing computational challenges in models like RT-2.
• RT-Trajectory overlays 2D visuals on training videos, boosting a robot's intuitive task learning. Success rates jump to 63% from 29% with RT-2.

McAfee Launches Project Mockingbird: AI-Driven DeepFake Audio Detection

• McAfee's Project Mockingbird: 90% accurate AI detects and combats deepfake audio threats, using contextual, behavioral, and categorical models against scams.

• Beyond scams, Project Mockingbird aids users in distinguishing between authentic and manipulated content, providing insights akin to a weather forecast for informed decisions.

• McAfee's December 2023 Deepfakes Survey reveals 84% of Americans are concerned about deepfake use. Worries include election influence, media trust erosion, impersonation, scams, cyberbullying, and explicit content creation.

• As pivotal election years loom, Project Mockingbird emerges as a crucial defense against AI-generated threats, aiming to enhance protection and safeguard democratic processes from potential consequences of manipulated audio and video content.

Check out AI Solution Models on  - https://www.dimensionless.ai/

Also, Explore In-depth articles written by us - https://blogs.dimensionless.ai/

Like, Share and Subscribe to our Weekly Newsletter for Top Tech Trends!!!

Comment below and tell us what you would like to know about the Tech-Industry from us.