It’s time to get into some big ideas.
I’ve mentioned Apple’s move into healthcare with the AirPods Pro 2 update. Matt Stoller of Big walks through history. This development stems from a 2017 congressional act that allowed over-the-counter hearing aids, aiming to break up a market dominated by a few firms and reduce costs. Previously, hearing aids were expensive, costing $3,000-$5,000 per ear, partly due to FDA regulations requiring prescriptions. The market was controlled by six major firms that integrated vertically, managing insurance plans and owning networks of audiologists1.
The Over-the-Counter Hearing Aid Act, led by Senators Elizabeth Warren and Chuck Grassley and signed by President Trump, mandated the FDA to allow hearing aids without prescriptions. However, the FDA initially delayed implementation until pressured by the Biden administration.
This regulatory change opened up the market, allowing companies like Apple to innovate.
TechNative asks, “How will AI change the future of IT Support?” AI is set to transform IT support by enhancing efficiency rather than replacing human roles. It addresses challenges like skill gaps and rapid tech changes by automating routine tasks and providing smart assistance. Successful implementation requires focusing on tasks suited for AI, training staff on new tools, and ensuring human oversight to maintain service quality. Companies that effectively integrate AI will likely achieve higher customer satisfaction and operational efficiency.
Tanay Jaipura, with another great article, highlights how the cost of intelligence in AI is rapidly decreasing, with GPT-4 equivalent intelligence dropping 240 times in 18 months. Key factors include increased competition among model developers, improved efficiency in compute, the rise of smaller and smarter models, new model architectures, and the shift towards on-device inference. As costs continue to decline, new applications and industries are likely to emerge, prompting founders to focus on innovative use cases that become feasible with lower intelligence costs.
And MIT Technology Review explains why OpenAI’s new reasoning model is a big deal. It ranks in the 89th percentile on Codeforces competitive coding questions, Among the top 500 high school students in the USA Math Olympiad, with 83.3% accuracy on math Olympiad questions (vs. 13.4% for GPT-4o). It has 78% accuracy on PhD-level questions (vs 69.7% from human experts and 56.1% from GPT-4o). The model uses “chain-of-thought” reasoning, potentially making it valuable for fields like drug discovery, materials science, and physics. However, comparisons to human-level skills should be approached cautiously, and measuring true reasoning ability remains challenging.
Why do we care?
Just a note – my own doctor wasn’t aware of the Apple moves into healthcare, and considered it a big deal. Pass that one on.
Whether it’s integrating consumer tech into healthcare, deploying AI to enhance IT support, or leveraging next-generation reasoning models, the key is to maintain a strategic approach—balancing innovation with caution and ensuring that new technologies are implemented thoughtfully and responsibly.