Apple swung for the fences by trying to reclaim AI as “Apple Intelligence” at its __annual WWDC __on June 10. I don’t know if the company will succeed with the branding play, but I do know that OpenAI hit a home run on Apple’s turf. That’s because Apple’s Siri voice assistant is getting a lifeline from OpenAI’s ChatGPT.
At WWDC, Apple __announced __that the company is partnering with OpenAI to allow Siri to supplement its responses with information from ChatGPT’s knowledge base after obtaining user consent. This will be effective with the rollout of iOS 18.
In addition, Siri is getting a number of major AI upgrades. For instance, Siri now has better contextual awareness, allowing it to link commands and understand the context of queries better based on what’s displayed on the screen and the user’s personal data. Siri can assist with writing tasks like emails, documents, and messages by offering proofreading, rewriting suggestions, tone adjustments, and smart reply options.
For certain complex Siri requests, Apple will prompt users to connect with ChatGPT for more complete responses. Additionally, Apple is integrating ChatGPT with other areas of the operating system, such as text composition assistance. (According to Apple, users will have control over when and if they choose to use ChatGPT.)
For example, if a user asks Siri, “Can you summarize the latest research on climate change and its economic impacts?” Siri might recognize that this request requires a more detailed and nuanced response. Instead of providing a brief or potentially insufficient answer, Siri will prompt the user to connect with ChatGPT for a more comprehensive response. Siri will analyze the complexity of the query. If it falls beyond the scope of standard pre-programmed responses, Siri will suggest, “Would you like to get a detailed answer from ChatGPT?”
Or let’s say that a user needs to respond to a complex message. They might say, “Siri, help me reply to this message explaining our new product features.” ChatGPT can draft a response that is detailed and tailored to the specific context. Within the Messages app, Siri can invoke ChatGPT to generate responses that are contextually accurate and linguistically refined. Users can review and modify these responses before sending them.
The integration extends beyond text-based interactions. If you ask Siri a question about a photo, document, or PDF, it can now connect to ChatGPT to analyze and provide relevant information. For example, you could take a picture of a plant and ask Siri to identify it or point to a paragraph in a document and ask for a summary.
The development is part of the larger roll-out of “Apple Intelligence,” a comprehensive software update for its devices that promises to deliver a personalized generative artificial intelligence experience to users. (Oh -- and did someone say, “obvious branding play for Apple”?)
This is all important because the more conversational AI becomes, the more AI can help people solve complex problems more naturally beyond the prompt-and-response format we use with text. (It’s worth noting that the latest version of OpenAI’s ChatGPT generative AI tool is incorporating a voice-based interface, too.)
For example, consider contextual understanding. Natural conversation allows AI to maintain context over multiple turns, understanding the nuances and references made earlier in the conversation. This leads to more coherent and relevant interactions. AI can provide personalized responses by remembering user preferences and previous interactions, creating a tailored experience.
Or consider complex query handling. A conversational AI can break down complex problems into manageable parts and provide step-by-step guidance. This is particularly useful in areas like technical support, planning, and learning. Users can ask follow-up questions, request clarifications, and refine their queries on the go, leading to more accurate and satisfactory outcomes.
Ironically, making generative AI more conversational is what voice search was supposed to be doing all along. Natural conversation eliminates the need for rigid commands or specific prompts, making technology more accessible and intuitive for everyone, regardless of technical expertise. Natural conversation with AI could bridge the gap between humans and machines, making AI a more organic part of our lives to the point where we won’t be writing articles like this one that examine the mechanics of how we collaborate with generative AI. When that happens, AI achieves an essential milestone: being pervasive and transparent.
Generative AI has gained immense popularity, largely due to how OpenAI has effectively built its brand around this technology. But the company has stumbled at times with its business relationships. Content creators, news sources, and entertainment brands have accused the company of exploiting their content to train ChatGPT. OpenAI cannot build a brand by burning bridges any more than Apple could succeed by isolating the iPhone from developers and partners, which, of course, it did not.
However, OpenAI has recently been extending its reach through better partnerships. In May, OpenAI and Reddit announced a partnership involving data access and AI-powered feature collaboration, with OpenAI also becoming a Reddit advertising partner. This enables OpenAI to use Reddit’s platform and user base for targeted advertising. The same month, OpenAI and News Corp. revealed a multi-year agreement allowing OpenAI to display content from News Corp.-owned outlets within its ChatGPT chatbot. This was followed by partnerships with Vox Media and The Atlantic.
Additionally, PwC has become the largest customer and first reseller of OpenAI’s enterprise product, ChatGPT Enterprise, solidifying OpenAI’s presence in the workplace.
The race to dominate the generative AI market hinges on the strength of ecosystems. Brands with ecosystems become pervasive, and when a brand is pervasive, it becomes a more natural part of our lives, as Apple’s iOS has.
The relationship with Apple has potential concerns about conflicts because of OpenAI’s own developments in voice technology for ChatGPT. But OpenAI’s voice integration for ChatGPT and its collaboration with Apple can be seen as distinct, if overlapping, product offerings. The Siri integration could be tailored specifically for Apple’s ecosystem, using ChatGPT’s capabilities in a way that aligns with Siri’s existing functions and Apple’s privacy standards. Meanwhile, OpenAI can continue to develop and enhance its own ChatGPT voice features independently.
Siri, which follows Google Assistant and Amazon Alexa in popularity, now enjoys the brand cachet of ChatGPT.
Bottom line: Apple needed to play catch-up with AI. The company just got some instant offense from OpenAI.