Google Is Blurring the Line Between Humans and Software

Imagine a future in which you have long-running personal conversations with the software that controls your home entertainment system while letting an algorithmic chatbot handle texting your mom for you. Google pretty much showcased that vision of the future at its annual developer conference, I/O, which kicked off Wednesday in Mountain View, California. 

In his first I/O appearance since he was named CEO last August, Sundar Pichai opened by introducing Google Assistant, a voice-activated virtual assistant akin to Apple’s Siri and Amazon’s Alexa. Like Siri, Google Assistant will be accessible to users through Google’s Android phones and tablets; like Alexa, she’ll also live in a stylish new standalone wireless speaker, which will go on sale later this year. (Evidently, tech companies can’t help but make all their assistants female: In the demos, Google Assistant speaks by default in a woman’s voice, like both Alexa and Siri.) 

“We want to be there for our users asking them, ‘Hi, how can I help?'” Pichai told an audience of developers and journalists. “Think of it as a conversational assistant. We want users to have an ongoing two-way dialogue with Google.” 

Google Assistant embodies Google’s vision for how people will interact with the tech giant in the future–through voice, ambient computing, and context-aware artificial intelligence. It is also a key feature of the company’s much anticipated Google Home device, also revealed at today’s event, which will compete directly with Amazon’s successful Echo. “Credit to the team at Amazon for creating a lot of excitement in this space,” Pichai said, while claiming that Google Home will be considerably superior in its “far-field voice recognition” (ability to decipher speech against background noise at different distances), its integrations with other smart-home systems like Google-owned Nest, and its aptitude to understand and manage things like restaurant reservations, travel itineraries, and contact lists. 

“It’s like having a voice-activated remote control to the real world whenever you need it,” Mario Queiroz, Google vice president of product management, said. According to the company, the same artificial intelligence that allows Google Assistant to understand conversational questions from humans also allows it to generate conversational snippets that human users can pass off as their own.

After Pichai finished showing off Google Home, he handed over the stage to Erik Kay, an engineering director who unveiled a new “smart messaging app” called Allo. 

In the crowded messaging market, Google is hoping to set Allo apart with a range of features: end-to-end encryption (like WhatsApp), disappearing conversations (like Confide), and a “Whisper/Shout” slider that lets you adjust the font size on each message.

But the “smart” part of Allo lies in its ability to auto-suggest quick text replies to incoming messages–not just generic ones like “I’m busy” but specific, context-aware answers that seem expressly designed to fool the person on the other end of the exchange into thinking they came from a person, not a bot. Demonstrating this feature, Kay showed an image of a text message thread in which an Allo user receives a friend’s dog photo. The suggested replies: “Cute dog!” “Aww!” and “Nice Bernese mountain dog.” 

Technology that frees you up from talking to people so you can spend more time talking to machines. That’s one form of progress.