For organized industries, AWS is from AI AWS with a safe and automated agent automation

Cook said that some customers allowed AWS to review their data and documents used to suspend the answers as correct or wrong, and found that the work created by the tool is being implemented similar to humans with a copy of the book of the rules in front of them. He added that the concept of truth or correct can often be subject to interpretation. Automated thinking does not have the same problem.

He said: “It was really surprising! It is surprising that people with logical backgrounds are in an internal communication channel arguing about what is true or not, and in five or six messages indicating the tool and realizing Oh, this is true.”

AWS has added new features to the optimum thinking exams for public version. These include:

• Support for adding large documents up to 80 thousand icons or up to 100 pages
• Verify the health of the most simple policy by preserving health verification tests for frequent employment
• The generation of automated scenario from prior definitions
• Natural language suggestions for political comments
• Customized health verification sets

Cook said that automated thinking verification operations are validated by the truth or right in the artificial intelligence system by proving that the model has not been shared with a solution or response. This means that it can provide organizers and institutions subject to concern that Non -inevitable nature of AI It can restore incorrect responses more confidence.

Ai nervousness and prove the truth

Cook put an idea that automated thinking examination helps to prove many AI’s nervous concepts.

AI Neurosymbolic It indicates the plural One of the nerve networks used by language models, with organized and logical thinking of symbolic intelligence. When nerve networks recognize patterns of data, AI uses clear rules and logical problems. Basic models often depend on nervous networks or deep learning, but because the models build their responses to patterns, they are so VulnerableAnd anxiety is still related to institutions. But Amnesty International is not very flexible without manual instructions.

Promise sounds in artificial intelligence, such as Gary Marcus said The nervous AI is very important to artificial general intelligence.

Cooking and AWS are excited Bring ideas from the nervous AI To the institution. Matt Marshall of Venturebeat spoke about AWS’s focus on methods such as automated thinking and combining mathematics and logic to artificial intelligence to reduce hallucinations In podcast.

Currently, a few companies are offered in reverse Amnesty International. These include Kognitosand Franz Inc. and Omini.

Bring mathematics to verify health

Automated thinking by applying sports proofs to models in response to inquiries.

It is used as a method called Modulo for contentment, as symbols contain pre -defined meanings, and they solve problems that involve both logic (then, then, or) and mathematics. Automated thinking takes this method and applies it to responses through a model and verify this against a set of policy or truth data without the need to test the answer several times.

For example, in preparing the institution, they want to prove that financial audit is correct. The model responds that the report contains unaccounted payments. Automated thinking tests divide this into a logical series:

(Forll ((R Report))

(=> (Contains unaprovidorpayments r)

(You must inspect P))))))

Then he moves to the definitions, variables and species that the user has identified on the foundation handrails and replaces the equation to prove that the model was correctly responded and based on the truth.

Make the agents greatly correct

Cook said that fake use cases can benefit from automated thinking examination, and that giving more access to the feature through the basis can prove its benefit. But he warned that automated thinking, and other AI nerve techniques, are still in its very early stages.

“I think it will have an effect on Aicenc AI, however, of course, the work is very speculative at the present time,” said Cook. “There are many techniques such as discovering mystery in the statement, then you find a kind of main delta between potential translations, then return to you and get refined on it, I think, it will be key in terms of the emotional journey that I saw through them began to play with artificial intelligence two years ago.”

Daily visions about business use cases with VB daily

If you want to persuade your boss at work, you have covered VB Daily. We give you the internal journalistic precedence over what companies do with obstetric artificial intelligence, from organizational transformations to practical publishing operations, so that you can share visions of the maximum return on investment.

Read with us privacy policy

Thanks for subscribing. Check more VB news bulletins here.

An error occurred.