This week in AI: Apple delays AI features launch, Taco Bell rolls out Voice AI in drive-thrus, EU AI Act takes effect

The launch of Apple’s (NASDAQ: AAPL) artificial intelligence (AI) features, dubbed “Apple Intelligence,” has been delayed from its expected September release to October due to concerns over stability and the need for more testing.

Initially planned to launch with iOS 18, which will come to iPhone and iPad in September, the features are now expected to arrive with iOS 18.1 and iPadOS 18.1, a few weeks after the debut of the latest OS versions. Regardless of the delay, Apple plans to give developers early access to Apple Intelligence this week through the iOS 18.1 beta version, allowing them to build and experiment with it.

Even when Apple Intelligence launches in October, it will be a watered-down version compared to what Apple originally intended. Instead of a full-featured AI suite—including transcription for phone calls, AI photo retouching and improvements in Siri’s natural conversation flow—only the AI features related to Siri will be live initially. Other AI features are expected to roll out in the first half of 2025.

Apple previously tried its hand at AI with its voice assistant Siri in 2011. Thirteen years later, Siri remains largely unpopular. Until Apple Intelligence launches, it’s uncertain whether it will be a desirable product. Apple has a history of products that sounded enticing on paper, like the Apple Vision Pro, that failed to live up to the hype come launch day.

AI-powered drive-thrus: Taco Bell’s voice AI

Yum! Brands, the parent company of Taco Bell, announced an expansion of its Voice (Audio) AI across drive-thru locations in the U.S., aiming to have the technology in hundreds of stores by the end of 2024.

“We’re expanding and accelerating our AI capabilities like Voice AI to deliver leading-edge technologies to our franchisees and to enhance the consumer and team member experience,” said Lawrence Kim, Chief Innovation Officer, Yum! Brands.

Instead of an employee taking your order at the drive-thru, customers will be serviced by AI. Taco Bell currently has these AI systems operational in more than 100 drive-thrus across 13 states, finetuning its models over the past two years.

The company claims its AI drive-thru improves order accuracy, provides a consistent, friendly experience and reduces wait times. This reduces the workload on employees, allowing them to focus on tasks that are more difficult to automate.

“Innovation is ingrained in our DNA at Taco Bell, and we view Voice AI as a means to improve the team member and consumer experiences,” said Dane Mathews, Chief Digital & Technology Officer, Taco Bell. “Tapping into AI gives us the ability to ease team members’ workloads, freeing them to focus on front-of-house hospitality. It also enables us to unlock new and meaningful ways to engage with our customers.”

Among the pie-in-the-sky narratives we typically hear about the technology, AI in fast food drive-thru workflows is practical. With a limited menu and a static process, ordering food at a drive-thru is ideal for automation and reducing human errors. It wouldn’t be surprising to see more fast food chains adopt similar AI systems to experience the benefits that Taco Bell says come with Voice AI.

AI wearables: A new approach to AI-powered companionship

A new AI wearable is preparing to enter the market, but instead of aiming to increase productivity, act as a personal assistant, or replace your smartphone, this device, named “Friend,” is designed to act as a companion and conversational partner. The Friend team believes that constant companionship is one of AI’s killer use cases and says many people currently use AI models to talk about their day and feelings.

Friend can be worn as a necklace or clipped to your clothes and works by continuously listening to the world around you. It interfaces with the user through a text-message-like app, providing responses based on what it picks up or what is said directly to it via a walkie-talkie-like button.

However, it’s questionable whether this is truly what people want. Recent AI wearables have generally flopped and have received more negative reviews than positive ones. Friend’s focus on companionship, rather than being a smartphone replacement, is a different approach, but it remains to be seen if AI can genuinely replace human connection.

Those seeking AI companionship likely already use existing AI models to accomplish that goal. Even if Friend’s distinction is its constant presence, the fact that it is “always with them” likely means the wearer is stepping out into the world into settings that are more social than being secluded, which probably means that the gap Friend is aiming to fill is not as large as they may have imagined it to be.

The EU AI Act takes effect

The EU’s AI Act went into effect on Thursday, August 1. Now, companies operating in the EU must comply with the Act and meet its reporting deadlines. The Act focuses heavily on protecting citizen rights, categorizing AI applications based on their risk levels. Lower-risk applications, such as spam filters and content recommendation systems, are subject to minimal regulations and must disclose their use of AI for transparency.

High-risk AI systems, especially those in sensitive sectors like healthcare, education and public services, face stringent regulatory requirements, including detailed documentation and mandatory human oversight.

The higher the risk to Europeans’ health or rights, the stricter the requirements to protect individuals from harm. Most companies have until 2026 to comply, but some AI models, like OpenAI’s ChatGPT, have 12 months to comply, while other high-risk applications have six months.

The EU is a historically challenging location for data-intensive companies due to its resident-first stance on personal data and privacy. Companies already operating in the EU must now comply with the AI Act, but it will be interesting to see how companies currently outside the EU navigate these regulations or if they choose to suspend their service offerings, similar to how Meta (NASDAQ: META) recently had to suspend its service in Brazil once its compliance with the local laws was called into question, or even make the decision not to enter the EU market due to these new requirements.

In order for artificial intelligence (AI) to work right within the law and thrive in the face of growing challenges, it needs to integrate an enterprise blockchain system that ensures data input quality and ownership—allowing it to keep data safe while also guaranteeing the immutability of data. Check out CoinGeek’s coverage on this emerging tech to learn more why Enterprise blockchain will be the backbone of AI.

Watch: Why AI & blockchain are powerhouses of technology

New to blockchain? Check out CoinGeek’s Blockchain for Beginners section, the ultimate resource guide to learn more about blockchain technology.