We believe that the integration between Artificial Intelligence and Blockchain will enable a new generation of Smart...

>We believe that the integration between Artificial Intelligence and Blockchain will enable a new generation of Smart Contracts.
>We have developed Websensors-Dapp for iExec as a solution for Artificial Intelligence-based Oracles and the results are promising!
>Moreover, any other event analysis platform or machine learning framework can (and should) be evaluated for this task.
>You can train your own Artificial Intelligence-based Oracle. We have developed some tools that can help you with this task
>enable a new generation of Smart Contracts.
>enable
>a new
>generation of Smart Contracts

medium.com/artificial-intelligence-for-blockchain-smart/how-to-use-machine-learning-algorithms-as-oracles-in-smart-contracts-238c6353526a

Attached: websense.png (1579x889, 370K)

Other urls found in this thread:

github.com/rmarcacini/websensors-iexec
dapps.iex.ec/dapp/rmarcacini/websensorsdoracle
software.intel.com/en-us/articles/ecosystem-support-for-eea-trusted-compute-specification-v10-improves-blockchain-privacy-and
vox.com/future-perfect/2019/4/8/18297410/ai-tesla-self-driving-cars-adversarial-machine-learning).
medium.com/iex-ec/why-your-dapp-needs-a-decentralized-oracle-2f2403f9fd7
twitter.com/NSFWRedditImage

that's nice but chainlink does it better

>KYC nodes
>no AI based oracles
OH NO NO NO

Great buzzword salad, but anyone who's spent 30 minutes looking into AI safety knows how easy it is to manufacture adversarial inputs for the current archetype of AI systems. You literally take the training regimen and turn it around so that instead of optimizing the model based on inputs, you optimize the inputs based on the model. For the brainlets: making an "optical illusion" etc to fool AI is really easy. And this is what we're gonna trust to trigger high value smart contracts? I dont mean this as the typical shithead fud, this is a really interesting and challenging problem to address. Same shit is going to be a problem for safety of self-driving cars when an extremist dickhead starts putting translucent overlays on street signs that make your car swerve, brake unexpectedly etc.

>look at me I think I'm smarter than a team of PHD's and the entire websensors team

Attached: 1534803314085.jpg (306x306, 20K)

Ok how about you try to refute any aspect of what I said?

None of what you said is actually true. It seems you don't know how AI works either. You can train AI to ignore inputs, so your "optical illusion" attempts would be for nothing. Also, there will be a reputation/staking system attached to this due to it working within the iExec platform so good luck trying to do any of what you said.

All in all you think you're smarter than a bunch of literal geniuses and it shows how much of a brainlet you are when you say dumb shit like

>Same shit is going to be a problem for safety of self-driving cars when an extremist dickhead starts putting translucent overlays on street signs that make your car swerve, brake unexpectedly etc.

because you don't understand how AI or machine learning works and the steps you can take to combat against this type of terrorism.

obviously it is not going to be used for high value smart contracts

>We believe that the integration between Artificial Intelligence and Blockchain will enable a new generation of Smart Contracts.

cool belief, build it!

the websensors project is kind of an interesting, early stage non-commercialized project that has nothing to do with iexec.
the iexec integration / "muh AI Oracles" is a toy proof of concept, built by one guy.
github.com/rmarcacini/websensors-iexec
overall, this is currently basement tier and has many things wrong with it.

Time will tell.
It's like buying ETH before ETH became ETH. You're taking a decent risk for potentially massive gains.
It could definitely be used for high value smart contracts once the systems are all built out and tested fully.

You guys don't seem to understand that this is still below-ground level stuff here. Imagine buying ETH while Vitalik was still in his dad's basement coding it.

>early stage non-commercialized project that has nothing to do with iexec.
Everyone laugh at this retard that hasn't been keeping up with iExec but thinks he knows everything

>dapps.iex.ec/dapp/rmarcacini/websensorsdoracle
>dapps.iex.ec/dapp/rmarcacini/websensorsdoracle
>dapps.iex.ec/dapp/rmarcacini/websensorsdoracle
>dapps.iex.ec/dapp/rmarcacini/websensorsdoracle

it's over. ETH and LINK already work and are known. by the time RLC has anything comparable it'll be too late. your shitcoin lost. adopt the standard or live in regret forever

No, faggot.
You already had a thread begging for people to buy into muh blockchain ai bullshit earlier today. It was funny. We all had a laugh. Now it is just sad.

Sage

RLC worked before LINK you retard kek
RLC was the first working decentralized oracle, look it up.
ETH is a shitcoin creator. Can't do anything RLC can do. Your fud is trash and so are you!

0/10 fud try again faggot

iexec is actually eth standard:
software.intel.com/en-us/articles/ecosystem-support-for-eea-trusted-compute-specification-v10-improves-blockchain-privacy-and

lol your own "AI oracles" example uses the eth blockchain to store its output (pic related)

Attached: Screen Shot 2019-07-12 at 9.33.00 PM.png (1416x352, 100K)

Note how chainlink is the designated oracle component of that standard...

because they are on eth, and with V4 they will have sidechain

link is not mentioned in that article

sorry it is as a c tier worth contributor

Attached: Screenshot_2019-07-13-07-42-40-767_com.google.android.apps.docs.png (720x1440, 273K)

i'm aware user. you seem more reasonable than OP. just wanted to save any unwary anons from getting too hyped about ai oracles

just because some iexecuck edited the document doesn't mean they contributed to the project like chainlink

kek nice cope. but go read the document now

Sidechain, retard. ETH will not matter when it comes to what RLC can do in the near future.

Ok this is pathetic. It could not be any more obvious that you're spewing this straight out of your ass. "Train the AI to ignore inputs" You clearly dont even begin to understand the problem of adversarial inputs, which arises specifically due to the ability to optimize inputs *based on the model's training regimen*. Second, hurr durr hey guys we'll use an AI to trigger this financial contract that cant be reversed but don't worry it will ignore any data that I don't find convenient. Third, staking and reputation have nothing to do with addressing this problem, because it's a data feed problem not an oracle integrity problem. Fourth, you keep using appeal to authority as a basis to refute my argument, as if this literal who team is infallible and "AI based" blockchain buzzword scams aren't a dime a dozen. Honestly I would love to be wrong here because it would be amazing technology if this works. But since you're so sure this massive problem can be easily avoided, why dont you explain to us brainlets just how you "train an AI to ignore inputs" to avoid this problem?

>spoonfeed me the answer because I still think I'm smarter than a team of PHD's in computing and a team dedicated to making an oracle dApp as well as Elon Musk who is able to tackle this problem in a centralized way.

>Honestly I would love to be wrong here because it would be amazing technology if this works
You are wrong. DYOR. I'm not here to spoonfeed you, only mock your inability to look for information as I buy more computation tokens of the future. Enjoy buying my bags at $400 per RLC in 3 years.

powerful.

Attached: the_next_currency.jpg (1888x1562, 390K)

as far as i know you cant specifically train ai to ignore aome inputs
however if smart contract will be trigerred by deterministic outcome ai wont be used for some expensive financial contracts or even cheap financial contracts because it simply does not need ai in that case.

>We believe that the integration between Artificial Intelligence and Blockchain will enable a new generation of Smart Contracts.

Buzz words. Don’t trust anyone trying to sell a product with AI in.

Even the big companies pushing AI eg self driving cars are BSing.

It cannot currently be done. It’s just buggy code actual intelligences (eg humans) can easily break.

>Even the big companies pushing AI eg self driving cars are BSing.
>It cannot currently be done.
Yet Tesla's are being improved day in and day out by AI.
All of you brainlets on biz aren't going to make it. I can't wait for the pink wojaks.

Oh yeah nobody is going to see through "hurr durr I wont spoonfed you" as your complete inability to address any of my points. Citing Elon Musk (yet another failed attempt to appeal to authority) just screams how ignorant you are about this issue, as Tesla's are wildly prone to manipulation in this exact manner (example: vox.com/future-perfect/2019/4/8/18297410/ai-tesla-self-driving-cars-adversarial-machine-learning). Also I hold RLC bags retard because I can see the value of decentralized computing. But "AI oracles" are a fundamentally flawed gimmick at this stage, and you shilling them like a desperate pajeet is not going to add any credibility or interest for RLC which is already a solid project on its own.

If you even read that article you would know that anyone dealing with AI are taking steps to defend against attacks. It's both ignorant and disingenuous to think that neither the iExec or Websensors team can come up with or already have potential solutions ready. AI oracles are not a gimmick, they are just another example of why iExec is going to be one of the frontrunners of the upcoming AI/oracle investment shitstorm that is about to hit the cryptosphere.

Nonetheless, iExec is doing oracles both with and WITHOUT AI, so this argument is retarded.

“Yet Tesla's are being improved day in and day out by AI.”

I doubt it, I think they’ve hit a wall (literally?). Stuck in development because of insurmountable bugs. Tesla isn’t the only company researching self driving cars. It’s been years now. None of them are going to admit they are no closer to solving it.
Simple roads fine. Actual roads with complex situations no.

Currrnt AI is a dumb non sentient algorithm that isn’t as resourceful or adaptable as humans and can be easily gamed/tricked.

Using AI to generate smart contracts based on training them on news articles is like 2 intractable problems in one. Until they can show it working I wouldn’t invest.

One issue I can see is people generating fake news stories to game the oracle so the contract resolves in their favour.

their approach to Oracles is based on machine learning. Machine learning isn't always correct.
It's more reliable a network of people and the majority of them agreeing on the event (Chainlink)

Nah, it's just one of 2 ways that iExec can approach oracles. The other way is here.

>iExec DOracle relies on the iExec cloud executing the off-chain logic before returning certified results to the blockchain. Certification is achieved using an on chain consensus (iExec’s PoCo consensus protocol) in order to ensure that the results of a DOracle execution can be trusted for the on-chain settlement of off-chain events.

>medium.com/iex-ec/why-your-dapp-needs-a-decentralized-oracle-2f2403f9fd7

websensors they are a different tipe of oracles?Anyway the tech ---ON PAPER--- seems intersting. It's a shame this coin couldn't pull a x100 in the 2017 bull run
Immagine this in 2017

Yeah websensors is an AI based oracle powered by iExec.

AI and smart conract is currently buzzwords when mixed. like vegan and fair trade. Dont be an idiot who automatically thinks they are better when together. It barely makes sense or is relevent. Seriously if you are browzing Jow Forums dont be a huge fucking idiot. I know its a lot to ask.