Excuse me, maybe Your Project will work great with Vision.
But for everybody else, saying that : "Vision is enough for a Test" ??? - is a statement centered for One (Thin AI) application only.
I am working on Consciousness for AGI.
My projects are quite Aware and Conscious.
To me Vision is a later thing that I want to include in my AI (but can prove its capabilities without it).
Vision and Image Analysis is One of my Career Specialties (Factory Robotics Camera Image Recognition Coding in my past).
I have worked on Visual AI many times (Edge, Blob, Facial, Cam to 3D Interpolating)
- but this is clearly Recognition only.
I have not seen any 'Intelligent Eyes' that can make Intelligent Decisions and Planning
(not even in SciFi books) ... - have you ?
- It is a specific Tangent - and does not in itself prove Intelligence.
Ie: A BumbleBee can find by Sight, and successfully feed off of Flowers very well;
- Does this make it very Intelligent / Reasoning / Planner ?
Mantis Shrimp can see in 18 primary colors (10x as many Colors as Humans);
- does this make them Superior in Intelligence to Humans...?
Second, Mapping is not limited to sight. You (or Robot) can Map out a Room or Building with only Hands or Legs or Wheels-and-Bumpers or even Sonar.
So where are 'Eyes' or 'Vision' necessary Only for rating Intelligence Level ...?
- A Blind Man is clearly Intelligent and Conscious - which thus proves that your (Vision only / AGI Theory) is clearly flawed.
Also, Other People here are working on different sides of General AI, including Language and Speech, Planning, Conceptual, etc.
= Being Fair to All of the Participants (and Truly Judging AGI / Full AI) , I am suggesting Mixed Types of Test Problems (All Tests done by All Projects) - for Fairly Judging between Different kinds of Projects between Participants. So, everybody's Different kinds of AI Projects: Speech, Concepts, etc can be judged fairly between each other. I am sure there are some Great Speech AI and Conceptual and Planning AGIs - completely without any Vision capabilities.
So, a Project with All (or Most) of AGI capabilities would do better -than a Project limited to One or Few of the Capabilities (which would make it more Weak AI / Not true AGI anyway).
Ie: Not just Vision AI.
Ex: a Project that can Identify and organize Words, or Answer Questions from Context (like about Cars and Televisions), or maybe even Read and Summarize a Book ;
- are clearly more Intelligent than a Program that can only Visually Differentiate between a Television and a Car Image.
Vision / Image Analysis is a Specialty in AI - specifically defined as a Weak / Narrow Applied AI - NOT as AGI General AI.
From Wikipedia Definition:
Artificial General Intelligence (AGI) is the intelligence of a machine that could successfully perform any Intellectual task that a Human being can ...
There is Wide Agreement among Artificial Intelligence Researchers that intelligence is Required to do the following:
- reason, use strategy, solve puzzles, and make judgments under uncertainty;
- represent knowledge, including commonsense knowledge;
- communicate in natural language;
- and integrate all these skills towards common goals.
- Sorry I do not see Vision listed Anywhere in this Definition...