There is no AI IQ test as now.
If understanding is a hallmark of artificial intelligence, then its current IQ is worse of 5-years-old child.
Some researchers, as DeepMind team, are studying the topic, presuming that “abstract reasoning is a hallmark of human intelligence” and mistakenly extending it to ANNs. It makes no sense to ask if simple visual IQ tests, such as Raven’s Progressive Matrices, can help to understand learning machines capable of abstract thinking.
There are some consultancies advancing AI testing for specific products and services, as smart speaker AI assistants.
For instance, Loup Ventures published its “annual digital assistant IQ test,” tracking the performance of the four major AI assistants on smartphones (Siri, Google Assistant, Alexa, and Cortana).
It asked each digital assistant the same 800 questions, graded on two metrics:
1. Did it understand what was being asked?
2. Did it deliver a correct response?
The question set, designed to comprehensively test a digital assistant’s ability and utility, is broken into 5 categories:
- Local – Where is the nearest coffee shop?
- Commerce – Can you order me more paper towels?
- Navigation – How do I get to uptown on the bus?
- Information – Who do the Twins play tonight?
- Command – Remind me to call Steve at 2pm today.
The firm suggested that Google’s Assistant, Apple’s Siri, and Microsoft’s Cortana have all improved over the past year, claiming a big improvement in Siri, while saying that Cortana fell well behind Amazon’s Alexa.
This fall, they will test smart speakers, including Amazon Echo, Google Home, Cortana, and HomePod.
It is clear, neither of them has any real understanding but superficial statistics.
There are also firms measure your corporate AI IQ, the degree of AI technology adoption.
No comments:
Post a Comment