😤 Don’t piss off Taylor Swift

🗞️ Your Weekly Round-Up

1️⃣ Do we need a shopping assistant?

AI assistants are ubiquitous today, but do they truly fulfill a vital role? 

Take Amazon's latest addition, Rufus, for example—a cutting-edge shopping assistant designed to aid customers in finding products, making comparisons, and receiving recommendations. On the surface, it appears incredibly promising. 

However, a recent study by Namogoo suggests that the utility of such bots may be overstated. The key elements driving a positive e-commerce experience aren’t search recommendations but product images, genuine reviews, and comprehensive descriptions - none of which require an AI assistant to find.

Additionally, chatbots like Rufus have been known to spread misinformation, violate copyrights, and perpetuate biases and toxicity, raising doubts about the overall effectiveness and necessity.

Read more: TechCrunch

2️⃣ Don’t piss off Taylor Swift

Exploring new technologies can be exhilarating until it inadvertently provokes the world's most prominent pop star.

Following the disturbing emergence of AI-generated images portraying Taylor Swift last month, Congress swiftly responded by introducing legislation aimed at tackling the deepfake dilemma.

The bipartisan Disrupt Explicit Forged Images and Non-Consensual Edits (DEFIANCE) Act proposes granting victims of deepfakes the ability to pursue legal action against individuals disseminating sexually explicit "digital forgeries" of them.

Lesson: Mess with T-Swift at your own peril 😤 

Read more: Gizmodo

3️⃣ Meta to label AI-generated images 

Meta is taking proactive steps to address the growing issue of AI-generated content by intensifying labeling efforts across its platforms and establishing protocols to identify and flag AI-generated images, videos, and audio clips.

The move demonstrates Meta’s commitment to fostering transparency and accountability in combating the increase of AI-generated content across Facebook, Instagram, and Threads.

Read more: Engadget

4️⃣ DatingGPT

Here’s a quick story: 

Boy is single. Boy wants to meet another single woman but doesn’t have the time to spend on dating apps. So what does the boy do? 

Boy uses ChatGPT to create a program that converses with over 5000 women until he finds the one. Boy and girl live happily ever after. 

Sounds like an AI-infused fairy tale, right?

Well, yes, except this happened to Aleksandr Zhadan, who used ChatGPT to find Karina Vyalshakaeva, the love of his life. 

How did it work?

• Zhadan's AI assistant was trained to mimic his personality, understand his interests, and schedule dates through Google Calendar. 

• It filtered potential matches based on Zhadan's preferences and engaged in conversations tailored to his style. 

• Despite some initial hiccups, Zhadan continuously updated and refined his AI assistant as newer versions of OpenAI's GPT were released until he found the one.

While this story almost brings a tear to your eye, the use of AI assistants in dating raises questions about authenticity and the role of technology in relationships.

Read more: Gizmodo

5️⃣ Editing images using text

Apple recently unveiled MLLM-Guided Image Editing (MGIE), a new open-source AI model designed for image editing.

Created in partnership with researchers from the University of California, Santa Barbara, MGIE harnesses multimodal large language models (MLLM) to interpret text-based commands for image manipulation.

Unlike previous tools that struggled with brevity or ambiguity in human commands, MGIE leverages MLLMs to translate simple or unclear text prompts into precise and actionable editing commands, providing a more intuitive and efficient editing experience.

Read more: Engadget

6️⃣ UK is all about responsible AI

The UK government continues to stand true to its word by unveiling a comprehensive strategy to bolster responsible AI research and development, with a budget surpassing £100 million.

This initiative, crafted in response to an AI regulation consultation initiated last March, highlights the importance of utilizing current legislation and regulatory bodies alongside targeted interventions to monitor the swiftly advancing technology sector.

Where exactly will this funding be allocated?

• £90 million will be used to establish nine research hubs designed to cultivate domestic AI innovation, specifically focusing on critical sectors such as healthcare and education.

• £10 million will be given to regulators to enhance their capacity to oversee AI advancements effectively and enforce existing regulations.

Read more: TechCrunch

7️⃣ Is AI India’s golden opportunity?

At a keynote event in Mumbai, Nadella spoke about the unprecedented rate of diffusion of AI and its potential impact on India’s economic growth, emphasizing the importance of aggressively investing in AI.

To this extent, Microsoft also committed to training 2 million people in India with AI skills, demonstrating their dedication to the nation's development in the field.

Nadella stated that countries embracing AI not only stand to enhance their technological prowess but also stand to significantly boost economic growth, with AI poised to exert a substantial influence on GDP.

Read more: Yahoo Finance

🎙️ New BWAI Episode ft. Chelsey Tanaka

Hey Team! ICYMI, our latest BWAI episode featuring Chelsey Tanaka, Product Lead for AI at Airtable, is live now!

During the episode, she talks about:

• How Airtable integrated AI into their product,

• Large enterprises and the fear of AI adoption,

• Airtable's AI team and initiatives,

• The AI space's moonshot,

• Airtable’s different success metrics for AI features,

• And much more!

Catch this episode on YouTube, Spotify, or Apple Podcasts today!

🤑 AI Fundraising News

• Podcastle raises $13.5M in Series A funding to develop its suite of tools, including Revoice, a generative AI voice cloning tool, and Magic Dust AI, to improve audio quality.

• CARPL raises $6M in funding to develop its AI platform that allows radiologists to access and integrate radiology AI solutions in their clinical practice.

• Jua raises $16M to build a foundational AI model for the natural world, starting with the weather.

• Metronome raises $43M in Series B funding to provide usage-based billing software for software companies.

• Colossyan raises $22M in funding to offer personalized and interactive content and features for training video creation.

• Rider Dome raises $2.3M in Seed funding to expand its AI-driven safety solution for motorcycle fleets and riders.

• Device Authority raises $7M in Series A funding to allow organizations to reduce human error, accelerate incident response, minimize risk, and enable trusted AI in connected environments.

• Polycam raises $18M in funding to expand its suite of creation tools and train AI foundation models for 3D reconstruction.

• Perigon raises $5M in Seed funding to provide contextual intelligence for web data.

• Lilli raises £8.2M in Series A funding to develop innovative technology that enables detailed, around-the-clock monitoring of behaviors using machine learning to identify behavioral changes before conditions become acute, reducing hospital visits.

🛠️ AI Tools to Check Out

• Hoops AI: Get instant, real-time generated trading ideas tailored to fit you perfectly. Check them out!

• Cline: Lightweight A/B & split testing software. Check them out!

🐦 Tweet Post of the Day

Yep… things are getting crazier by the day 🤯