Apple readies device leasing plan as Google adds focused Gemini AI models

Apple is preparing a new way to pay for devices that could lower the upfront hit for iPhone buyers, Bloomberg News reported. For investors, the plan points to a familiar Apple strategy: make hardware easier to upgrade, then build more services and AI usage around each device.

Apple shares edged higher after Bloomberg reported that the company is working with buy now, pay later company Klarna on a device leasing program called Apple Upgrade, according to the CNBC Investing Club. Buy now, pay later services let customers spread payments over time instead of paying the full price at checkout.

Bloomberg reported that Apple Upgrade is expected to launch July 28 and cover most iPhone, Mac, iPad and Apple Watch models. The service will initially be available in the U.S. through Apple’s website and its physical stores, according to the report.

The setup would give customers a choice at the end of a lease term: move early into a newer model or keep the device. A lease is different from a traditional purchase because the customer pays for access to the product over a set period, rather than owning it outright from day one.

The CNBC Investing Club said the timing could help Apple address higher device prices tied to rising memory costs. Lower monthly payments may also appeal to customers who have kept an iPhone for four or five years, or longer, and have delayed upgrading.

That upgrade cycle matters for Apple’s AI push. The CNBC Investing Club noted that newer Apple hardware supports Apple Intelligence, the company’s AI offering, which is expected to get a makeover this fall. More powerful devices can also run a wider range of apps, giving Apple more ways to earn revenue after the initial hardware sale.

Google, meanwhile, announced three specialized Gemini models: Gemini 3.5 Flash-Lite, Gemini 3.5 Flash Cyber and Gemini 3.6 Flash. The new models are aimed at different use cases, a shift that reflects how companies are trying to control AI costs while still using the technology more widely.

A token is the basic unit of data an AI model processes. The more tokens a company uses, the higher its computing bill can become. The CNBC Investing Club said businesses are moving toward token efficiency after facing larger-than-expected AI compute costs, which makes smaller or more specialized models useful for tasks that do not require the most powerful system available.

Google said Gemini 3.5 Pro is being tested with partners and will become broadly available when it is ready. The company also said its team has started pre-training for Gemini 4, which it described as its most ambitious pre-training run yet.

The updates arrive before Alphabet’s earnings report on Wednesday evening. The CNBC Investing Club said investors are watching whether Alphabet, Amazon, Microsoft and Meta Platforms raise capital expenditure outlooks, meaning planned spending on assets such as data centers and computing infrastructure, while signaling more AI investment growth into 2027.

This story draws on original reporting from CNBC.