☕️ Figma On Thin Ice After Ripping Off Apple's Designs

Figma may have been a bit too gung-ho with its AI feature suite debut, being forced to temporarily pull the plug on its Make Designs feature after being accused of ripping off Apple's designs. Other key takeaways of the week include:

• Meta forced to suspend its AI training activities in Brazil by authorities

• Cloudflare debuts a system to help website owners combat AI content scrapers

• Google's emissions goal was dealt a hefty blow in a recent environmental report due to AI demands

Join us at AI Tangle as we untangle this week's happenings in AI!

Figma temporarily disables its Make Designs AI tool for copying Apple's Weather app design

With companies chucking out AI features and tools left, right, and center, Figma hopped on the bandwagon, too, recently, by revamping its UI and adding a suite of AI tools to go along with it, one of which ended up being a source of controversy - Make Designs. Make Designs, a beta AI feature that helps users quickly mock up apps using generative AI, was recently switched off following a social media post that made rounds showcasing the eerie similarities between Make Designs' results and Apple's Weather app layout.

​

How did this situation unfold?

Andy Allen, CEO of Not Boring Software, made an X/Twitter post where he accused Figma of heavily training its tool on existing popular apps. Figma CEO Dylan Field later replied about said removal, mainly pinning the blame on himself for strict deadlines and defending the company's attitude towards developing AI tools. However, in a later interview with Figma's CTO Kris Rasmussen, he couldn't say for sure that Make Designs wasn't trained on existing app designs as the underlying AI models behind the tool weren't developed by them, referring to OpenAI's GPT-4o and Amazon's Titan Image Generator G1. Rasmussen said Figma expects to re-enable Make Designs "soon" after ironing out its issues and adding more precautions.

Brazilian authorities slam the brakes on Meta's recently updated privacy policy regarding AI training

Brazil's National Data Protection Authority (ANPD) recently suspended Meta's new privacy policy for using personal data to train generative AI systems. The decision halts data processing across all Meta products and imposes a daily fine of 50,000 reais (around $8,836) for non-compliance, citing the risk of significant harm to fundamental rights and stating that Meta must revise its policy and confirm the suspension of data processing for AI training. Meta, unsurprisingly, isn't happy with the decision and is calling it a setback for innovation.

​

AI energy demands cause Google's carbon emissions to skyrocket by nearly 50%

Google's 2024 environmental report is out, and it revealed a nearly 50% surge in emissions since 2019, a massive blow to its net-zero emissions goal for 2030. Emissions rose 13% year-over-year in 2023, driven mainly by increased data center energy consumption and supply chain emissions as a result of ever-increasing AI demand. Despite challenges in implementing renewables at a grand scale, Google remains committed to reducing AI's environmental impact through efficient infrastructure and emissions reductions, even if its recent environmental report says otherwise.

​

Apple's App Store chief gets a seat on OpenAI's board following the iOS and ChatGPT integration partnership

Apple's App Store chief and former marketing head, Phil Schiller, is reported to have joined OpenAI's board as an observer to attend board meetings (but not vote) following the company's partnership with ChatGPT. Although ​Bloomberg reports​ that no money is currently being exchanged, it's expected that Apple may receive a percentage of ChatGPT subscriptions via its platforms. Having both Apple and Microsoft on the board could impact future plans, as it's quite unusual for Apple execs to take board seats at partner companies.

​

Cloudflare debuts a system to block AI web crawlers from scraping website content

Popular cloud service provider Cloudflare recently announced a new tool to combat AI scraping bots on websites hosted on its platform. Some AI vendors allow website owners to block bots using the robots.txt file, but not every AI scraper heeds these rules. To address this, Cloudflare analyzed AI bot and crawler traffic, upgrading its automatic bot detection models by attempting to detect whether a bot is trying to mimic human behavior. Cloudflare also now provides a form for hosts to report suspicious AI bots, manually blacklisting them over time.

​

YouTube adds a feature to make it easier to report and take down AI-generated videos that simulate your likeness

YouTube quietly rolled out a few changes to allow people to request the removal of AI-generated content that simulates their face or voice under YouTube's privacy request process. A takedown submission doesn't mean that the content will be removed instantly, however, as YouTube says it will make its own judgment about the complaint based on various factors. Once a complaint is submitted, the content's uploader will have 48 hours to make a move before YouTube will initiate a review.

​

Europe's latest $220 million growth-stage VC fund, Forestay, is looking to inject more cash flow into AI

Swiss venture capital firm Forestay closed its second fund, Forestay Capital II, at a hard cap of $220 million aiming to fund enterprise startups in both Europe and Israel with a sweet spot for leading growth rounds of $10-15 million. Forestay has so far backed 13 companies, including Scandit, K2view, and Wasabi. Led by Frederic Wohlwend, the former Global Chief Digital Officer of Merck KGaA and Serono, Forestay specializes in enterprise AI and SaaS investments. Forestay's emergence is especially valuable in Europe where growth-stage funding options are few and far between compared to the US.

​​Motiff - Motiff is an AI-powered design tool that fosters collaboration between humans and AI, allowing team members to design, align, and build in a single platform.

​Contentful - Create attention-grabbing digital experiences with the help of Conteful's AI-powered toolkit, allowing you to create websites and mobile apps to captivate every audience.

​Lunroo - Lunroo offers a complete suite of tools to help you grow your social media presence, engage with your audience, and drive more traffic across platforms like LinkedIn, TikTok, YouTube, and Instagram.

​Briefy - Briefy turns all kinds of lengthy content into concise, structured summaries and saves them in your knowledge base for later review.

Is AI Even Making Any Money? (2-min read)

A recent report from Goldman Sachs has some harsh words for the AI industry. Despite being ludicrously expensive to fund and develop, AI-based systems simply aren't bringing in any meaningful returns yet.

AI June Wrapped Up (33-min watch)

Fireship, a YouTube channel covering programming and AI developments in a comedic fashion, goes over the most significant happenings in the AI space in June to catch people up on the news.