Seven Ways You Interact With AI Before Lunch
6:00
Wake up, smartphone alarm rings
For the 80% of Americans who check their phone within a half-hour of waking up, AI is likely to be running on the device and in the cloud to help determine…
- When the screen backlight should turn on
- When to unlock the phone (biometrics like facial recognition, fingerprint, etc.)
- Recommendations (i.e. to check weather, directions, etc.)
- The voice assistant’s natural language processing and actions (i.e. Siri)
- Intelligent reminders based on prior activity
6:15
Peruse news, email or social media on smartphone
86% of Americans get their news from a smartphone or tablet, with nearly as many checking social media or email daily. AI is hard at work in each, working to…
- Surface relevant news stories from thousands of publications and hours of video
- Make social networks more engaging or useful, like how LinkedIn matches jobseekers with open positions or Twitter helps users rethink potentially abusive Tweets
- Target and serve ads based on billions of data points and >100 features, like weather, what a user clicked on in the past month and past purchases
- Increase relevance of content or search results while also filtering out hate speech, offensive images and illegal activity
- Make email safer by filtering out out 99.9% of email spam and scanning 300 billion email attachments for malware weekly
6:45
Breakfast
83% of Americans agree breakfast is the most important meal of the day, but they may not know that their morning cereal or fruit…
- Is probably healthier, fresher and more sustainable due to >$800M spent on AI in agriculture globally last year
- Is processed by intelligent farm equipment from the likes of John Deere that uses sensors and AI to increase yields and lower waste across >100M acres, >200 ML features
- Benefits from an intelligent supply chain that leverages AI to drive resilience and responsibility from farm to fork
- Is trucked to stores and restaurants via routes optimized by ML models
- Is restocked, picked or priced in a grocery store based on ML inferences, ultimately powering smoother operations and better customer experiences
7:00
Coffee run
Nearly half (48%) of Americans get their daily coffee fix from a quick service retailer or restaurant or store, where AI underpins…
- Inventory management and demand forecasting
- The in-app experience: customers who order ahead and collect rewards likely see personalized content and offers informed by ML models
- Credit card transactions, as credit card companies leverage models to mitigate and prevent fraud
- Whether a new coffee shop location opens closer to home, and whether it has a drive-through
- What practices and rules the establishment has in place due to COVID-19
7:10
Exercise, health or wellness time
57% of Americans are prioritizing their health and wellness this year. AI is often a silent partner in these efforts, powering…
- Fitness tracking devices; ML models help smart watches accurately track heart-rate and other vitals, while smartphones count steps and analyze sleep patterns
- Modern exercise machines; Tempo, for example, leverages computer vision to track form in real-time
- Healthcare; whether it’s whole new fields like precision medicine, improved cancer detection, a better patient experience through telemedicine or easier access to benefits, AI is revolutionizing many aspects of healthcare
- New drugs; ML models are routinely used in the biotechnology and pharmaceutical industries to aid in research, including most prominently in the development and distribution of Moderna and Pfizer’s COVID-19 vaccines
8:30
Commute into work
For the 86% of U.S. workers that commute into work by automobile, AI undergirds…
- Many of the advanced safety systems, like automatic emergency braking and forward collision warning, that come preinstalled on 92.7% of new cars sold. Together, these ML-powered systems reduce crashes by over 46%, on average
- Any assisted-driving or self-driving system. For example, Tesla Autopilot is powered by neural networks that involve 48 networks that take 70,000 GPU hours to train — together outputting 1,000 distinct tensors (predictions) at each timestep
- Smartphone apps that estimate commute times and optimize routes from billions of possibilities
9:00
Work
The 61% of Americans that comprise the U.S. workforce rely on AI running behind the scenes when they…
- Access the internet and work systems; internet service providers, telecommunications companies and cloud providers all leverage machine learning to optimize network performance and prevent outages
- Get paid; workers at over one in five global companies report that at least 5% of their company’s earnings before interest and taxes is attributable to AI
- Take a break to watch a video or do some online shopping; AI-powered personalization engines recommend content and products or target users later with offers to buy or watch more
- Ensure customers are happy; 90% of companies with top brand recognition and high levels of brand loyalty use AI solutions to increase customer satisfaction
In between when you wake up in the morning and when you have lunch, over 1,000 models have made decisions that touch your life. The small decisions these ML systems make have a massive impact on consumers and businesses. At Arize, we are building software to help teams troubleshoot the most complex systems ever built and provide guardrails when those systems make high stake decisions.
— Jason Lopatecki, CEO and co-founder of Arize AI