Apple Builds its own AI Chatbot to Reduce Google Reliance

After months of uncertainty, Apple has clarified its position in the AI domain: the technology giant is now venturing into the creation of its own AI-powered chatbot to enhance its product portfolio.
Bloomberg reports that Apple is assembling a team specifically designed to develop what it refers to as an "answer engine," similar to OpenAIās ChatGPT.
Since ChatGPT's launch in 2022, Apple has been attempting to catch up to its competitors, with its internal AI development team yet to create a comparable product to the leading generative AI tools.
The technology giant's performance, particularly in comparison to its peers Meta and Google, has been a point of frustration for investors.
This has driven Apple to consider partnerships with OpenAI or Anthropic to revamp Siri, but the financial implications were too steep to sustain such collaborations.
There was even consideration of acquiring a leading AI firm outright, with Perplexity, a popular chatbot company, being a potential target.
The shift towards in-house development marks a new chapter for Apple under Tim Cook and his leadership team.
Siri's limitations have sparked a rethink
Apple's current AI initiatives, under the label Apple Intelligence, primarily focus on efficient task execution, like summarising notifications and organising photos.
However, there's a noticeable deficiency in search capabilities.
This limitation becomes more apparent as Siri's capacity to handle basic queries consistently is questioned, often redirecting users to Google or handing over tasks to ChatGPT through a pared-down interface.
Devices devoid of screens, like the HomePod, highlight the necessity of more fluid voice interactions.
The tangible value of tools such as ChatGPT has altered any initial doubts Apple might have had.
Users employ these platforms for a variety of purposes, from solving complex problems to engaging in creative processes, often relying on them as their primary search engines.
The 'answers' team takes shape
Introducing Apple's "Answers, Knowledge and Information" (AKI) team, led by Senior Director Robby Walker.
Robby, who had oversight of Siri before facing challenges due to delayed implementations, is now leading this initiative with former Siri team members.
The team is developing what they refer to as an "answer engine" ā a solution designed to scan the web and provide responses to general queries.
Planned developments include:
- A possible standalone app offering direct interaction with users
- Upgraded infrastructure for future enhancements to Siri
- Potential integration with Spotlight and Safari's search tools
- Support for enriched information experiences throughout Apple's system
Job openings posted on Apple's career page indicate a broader scope for the project.
Recruitment priorities include individuals with "experience in search algorithms and engine development" to inform "the future of information connectivity".
Apple, Google and the shifting ground of web searches
Apple presently benefits from close to US$20bn annually through Google, which maintains its spot as the default search engine on Apple devices.
This lucrative arrangement, however, is under scrutiny from US regulators who may end the partnership due to antitrust concerns, potentially resulting in massive revenue losses for Apple.
By creating its own "answer engine," Apple aims to mitigate this impact by directing user interactions through its own technology.
Reports from Bloomberg suggest that Apple’s growth in services is intricately linked to the Google deal, heightening the importance of alternative income streams.
Concurrently, the search paradigm is swiftly changing.
Generative AI has broadened access to information retrieval, with tools like ChatGPT offering dynamic dialogue-based interactions that eclipse traditional keyword-focused searches.
Apple’s services head, Eddy Cue, has recognised in court that search powered by AI is pivotal to the future, highlighting potential partnerships with start-ups such as Perplexity.
Is an exodus of talent hampering Apple?
The search-sector expansion arrives amid a turbulent phase within Apple's AI departments.
The Foundation Models team, part of Apple's AI strategy focused on large language models for generative AI, has seen a wave of departures recently.
Four leading researchers have transitioned to Meta's new intelligence lab, including the architect of the team, Ruoming Pang, as well as multi-model development leader Bowen Zhang.
Meta's enticing salary offers, reportedly several times what Apple provides and the chance to engage with more advanced technologies appear to have been irresistible incentives.
These departures shed light on internal friction within Apple Intelligence, with criticisms aimed at the foundational models for falling short of expectations.
The long road ahead
Despite these ambitious planning stages, an eventual product launch is anticipated down the line, according to industry insight.
Apple is likely to prioritise privacy and coherence with its existing services over competitors' standalone chatbot approaches.
Based on historical patterns, any rollout will prioritise meticulous testing across a global array of devices before a public debut.
It is evident that Apple can no longer afford to remain passive with respect to AI innovation.
Presently, Apple’s pursuit isn't merely about entering the chatbot domain but catching up with its competitors, who have long since refined their technologies.
"In the competitive AI landscape, Apple is signalling it’s ready to reclaim control over its AI destiny," says Sarath Nair, Data & AI Director at Cognizant.
The dynamic realm of AI reveals that even Apple’s hallmark pursuit of perfection — a legacy from its iconic leader Steve Jobs — faces limits.



