Monday, December 23, 2024

AI Briefing: What would Steve Jobs say about Apple’s ambitions for AI?

Must read

Decades before the first iPhone, Apple co-founder Steve Jobs expressed his vision for the future of personal computing.

In a 1985 speech, Jobs imagined a world that now seems almost possible thanks to innovations in generative AI. Despite personal computing at the time being “in the tank,” he hoped it would someday have “tremendous momentum.” Even before the 1990s tech boom and bust, he predicted a future with “free intellectual energy” and an era where it would be possible to use computers and source material to ask questions to the world’s brightest minds — both past and present.

“My hope is someday when the next Aristotle is alive, we can capture the underlying worldview of that Aristotle in a computer,” Jobs said. “And someday some student will be able to not only read the words Aristotle wrote but ask Aristotle a question and get an answer. That’s what I hope that we can do. So this is a beginning I think.”

With Apple rolling out new AI features during last week’s WWDC 2024, it’s hard not to wonder if Jobs might actually feel his company has gotten quite close to that vision. (If only we could ask him now.)

The annual developer conference made plenty of headlines with various AI updates across Apple’s operating system and apps. While reaffirming its focus to customer-centric innovation, Apple replaced the usually ubiquitous generic mentions of AI with a more on-brand alternative called Apple Intelligence: “AI for the rest of us.”

Apple’s AI updates were across a range of products. Along with features for iPhones and MacBooks, and Apple Watches, the company also debuted major updates for software including Siri, mobile apps and newer platforms like the VisionPro. The company also introduced a new partnership with OpenAI to integrate ChatGPT into various parts of Apple’s operating systems.

Many of the updates won’t be available until later this year, but the big question will be who benefits most? Will the features help Apple catch up to earlier innovators like Google and Microsoft? Will OpenAI benefit from Apple’s hardware and brand? Or will consumers find new ways to interact with generative AI in their daily lives?

Thomas Husson, vp and principal analyst at Forrester, said consumer-facing brands haven’t been able to crack the code for generative AI. And while Husson thinks Apple isn’t yet an exception to that, he said WWDC once again introduced developers tools to improve customer experiences. He also noted major updates for Siri, which could spur innovation for voice AI on Apple devices.

“What matters is the total sum of improved experiences that brands will be able to develop on top of devices across Apple’s ecosystem and how consumers perceive them,” he said. “Apple resisted constantly referring to the overhyped gen-AI buzz and highlighted that they have been investing in many different AI technologies such as machine learning for quite some time.”

Improvements for Siri has agency innovation teams wondering what it could mean for the future of voice interfaces. It also might help weave together interoperability across Apple devices ranging from iPhones and MacBooks to CarPlay and VisionPro. Others note it could help spark new types of search optimization across devices and apps. 

“This sort of voice interaction model is actually the one thing that could be consistent across all of that hardware,” said Chris Weathers, vp of technology at VML. “It’s hard to predict the future but if you look at some of the crumbs that are in front of us now and try and see well, where do they potentially converge in a few years, it’s right there.”

Apple’s features might also enable some people’s first interactions with generative AI. According to a recent BCG global survey of 20,000 consumers, only 20% of people over the age of 35 had tried generative AI tools. Beyond just ChatGPT, one way people might interact with AI on Apple devices is through new feature called App Intents that lets Siri control. Mark Abraham, managing director and senior partner at Boston Consulting Group, said the updates could change not just how people use search engines, but also how they interact with in-app virtual assistants and search across numerous apps.

“Being able to inject these tools more easily into apps — and frankly I think even apps themselves will change — is one of the best trends we’re going to see,” said Abraham, who leads BCG’s North America marketing, sales and pricing practice. “…Where the unlock will be is if developers have access more easily through the technologies through OpenAI and other gen AI tools. I think Apple, like Microsoft, will make many bets here. And if they’re more easily able to integrate those into their apps, that’s what’s going to change.”

There are also questions about what Apple’s deal with OpenAI might compare with past deals such as its deal with Google as the default search engine, which is currently the focus of an antitrust trial with the U.S. Justice Department. Will Apple end up partnering with other AI providers in the future? According to a report by Bloomberg last week, Apple’s deal with OpenAI doesn’t include either company paying the other but in the future could include some sort of revenue share.

As Apple partners with companies including OpenAI, will privacy concerns and other safety issues with generative AI end up eroding the halo Apple has for years? It’s too soon to tell, but Apple made sure to point out new ways it’s thinking about privacy with its debut of Private Cloud Compute. 

Apple’s reputation for high brand trust among consumers will be crucial as it broadens its generative AI capabilities and partnerships, said Gartner analyst Nicole Greene. Apple emphasized its commitment to privacy and data protection and assured users its AI models running on their devices would remain on those devices with Apple having no access to the data. Greene noted that Apple also reinforced its stance by requiring businesses with cloud-based applications to undergo code audits to prove they are not capturing any data.

“Apple is not in the data monetization business as of now; they are focused on selling devices, so this decision aligns with their current business model,” Greene said. “An outstanding question is how that will apply to the relationship with OpenAI. They also said they would open up their hardware to others to validate that they’re doing, which is a game-changing and unique position, but how they’re going to monitor all of this is still an outstanding question.”

With a vast ecosystem of users, Apple is well positioned to successfully monetize its consumer base through integrating generative AI with hardware, Greene noted. And with so much content captured on phones, she thinks Apple could help scale the use of AI for content creation as well as authenticity and provenance.

“Apple is consumer focused (not enterprise) so how they choose to release gen AI capabilities across the experience will play a large role in cultural and societal adoption of the technology,” Greene said.

Apple’s approach to its users is quite different from companies like OpenAI, which often touts the capabilities of the AI models first and features powered by them second.  And as Apple and OpenAI work more closely together, will Apple’s philosophy of simplicity, privacy and ethics bleed into OpenAI’s own culture?

Even a modern Aristotle – human or not — can’t answer that question quite yet.

Prompts and Products: other AI news and announcements

  • Getty Images has a new deal with Picsart, a content creation platform, which will allow Picsart to develop a new AI image generation model trained on Getty Images’ licensed creative content. 
  • Adobe announced new features for its Adobe Express platform including expanded access for its Firefly 3 image model, ways for companies to use their own AI models, and new AI integrations with other major platforms like Microsoft Copilot and ChatGPT. 
  • Stability AI released a new AI image model called Stable Diffusion 3 Medium, which follows the release of Stable Diffusion 3 earlier this year.
  • Perplexity, an AI search startup, faced controversy over its new Pages features after journalists at Forbes accused it of using original reporting without citation. 
  • Amazon announced plans to invest $230 million in generative AI startups, with part of that funding coming in the form of AWS credits that can’t be transferred to other cloud providers.
  • Yahoo launched a new version of its news app powered by Artifact, a startup founded by the co-founders of Instagram that Yahoo acquired in April.
  • The AI video startup Luma AI debuted a new “Dream Machine” that lets people create realistic videos using text-based prompts.
  • Yext acquired Hearsay, a digital client services provider, with plans to integrate both for AI marketing efforts across various platforms.
  • The Tribeca Film Festival screened two separate series of AI-generated short films, with one in collaboration with Runway and the other with OpenAI’s Sora.
  • Ahead of Cannes Lions this week, Publicis Groupe released a new “BSBot” that it claims will let clients upload audio, images and text during the conference to see which companies’ AI offerings are “BS.” (If anyone uses the BSBot to test Publicis Groupe’s own offerings, feel free to get in touch.)
  • Microsoft said it will delay the rollout of its recently announced Recall tool, citing privacy and security concerns.
  • Spotify announced a new in-house creative agency that will use AI for voiceover ads.

1s and 0s: Reports and research

  • A new Forrester report found less than a third of U.S. B2C marketing executives surveyed said CMOs are leading their organization’s AI efforts.
  • A new report from BCG found more than 80% of a group of 200 CMOs surveyed said they’ve deployed AI tools to help with automation, speed, and productivity,” but more than 70% worry the tech will make their brand voice more bland.

AI stories from across Digiday

Latest article