SingularityNET is a decentralised AI marketplace which has the ultimate goal of forming the basis for the emergence of the world’s first true Artificial General Intelligence (AGI).

One of the brightest and most respected minds in AI leads the SingularityNET project, Dr Ben Goertzel.

“Current speed and cost issues with the Ethereum blockchain have increased the urgency of exploring alternatives for SingluarityNET’s blockchain underpinning,” says Goertzel.

“The ambitious Ethereum 2.0 design holds promise but the timing of rollout of different aspects of this next-generation Ethereum remains unclear, along with many of the practical particulars.“

SingularityNET claims that Cardano has now reached a level of maturity which makes it possible to port such a complex project to the new blockchain.

Back in August, Goertzel gave a talk at the Cardano Summit on why functional programming – enabled by Cardano’s Haskell and new Plutus languages – is invaluable for blockchain-based AI:

Current AIs in the SingularityNET marketplace are designed for specific, relatively straightforward tasks like image/language processing, time series analysis, and genomics data analysis. The project’s Android SDK has been used for tasks like separating vocals from music in the SongSplitter app.

While useful, these AIs show the current limitations of the technology today. SingularityNET, true to its name, has far more ambitious plans.

The broader vision of SingularityNET is to decentralise AI away from “big tech” and prevent AIs from being siloed so that one AI can outsource work to others and use their specific expertise to solve problems. This will ultimately bring us a step closer to AGI which acts more like a human by seeking help where needed.

In a blog post, SingularityNET provides further technical details about its decision to switch to Cardano:

“There may also be synergies between SingularityNET-Cardano integration and the OpenCog Hyperon initiative, which is focused on creating a more scalable, flexible and usable successor to the current OpenCog AGI R&D platform (which underlies a handful of specialized AI agents currently running on the SingularityNET network).

The OpenCog AGI design involves a metagraph knowledge store called the Atomspace, concurrently and cooperatively acted on by a number of different cognitive processes representing different learning and reasoning methods such as probabilistic logic, evolutionary learning, pattern mining and neural pattern recognition. Currently, to integrate OpenCog into SingularityNET, one creates a SingularityNET agent wrapping a whole OpenCog system with its own internal Atomspace and AI-process population.

However, in a SingularityNET-on-Cardano approach, it may eventually be possible to take a more decentralized approach in which the Hyperon Atomspace is provided as a service to any SingularityNET agent who needs it, and many of the cognitive processes involved in the Hyperon design are represented as SingularityNET agents that interact with Atomspace via channels set up via SingularityNET protocols. Such an approach would exploit the deep commonalities between the new version of OpenCog’s Atomese language being created for Hyperon and the dependent type based API of APIs under current exploration. The result would be a more fundamentally decentralized approach to AGI design.”

A fascinating interview between Cardano founder Charles Hoskinson and Dr Goertzel can be viewed here:

Cardano vs Ethereum

Ethereum is currently the world’s largest decentralised platform but suffers from slow speeds and increasing transaction costs. A switch to a more efficient and environmentally-friendly Proof-of-Stake consensus is underway which – along with new scaling innovations – should address Ethereum’s issues. However, it’s expected to be several years before Ethereum 2.0 is fully rolled out.

Cardano has observed Ethereum’s problems and is taking its time to address them with a scientific and peer-reviewed approach; which brings legitimacy to the project that will be needed for enterprise adoption.

While it could appear from the outside like Cardano has been lazy and is far behind other projects – after all, it’s yet to even support smart contracts – this is far from the case. Cardano is often ranked top of all blockchain projects for development activity and has continued signing large partnerships.

“Cardano has gone from strength to strength this year, and having the backing of such a prominent organisation only reaffirms this,” comments Hoskinson. 

“SingularityNET is a project we’ve followed for a long time, and we’re excited to see how the Cardano blockchain can help SingularityNET realise its ambitious goals.”   

Smart contracts are due to launch on Cardano in the coming months as part of what it calls its ‘Goguen’ phase. Unlike Ethereum, Cardano is using Proof-of-Stake from the start and won’t have the speed, cost, and scalability problems of the current decentralised platform leader.

Cardano will even become the most decentralised network in the space following the recent successful launch of its ‘Shelley’ upgrade.

On its website, Cardano explains:

“We expect Cardano to be 50-100 times more decentralized than other large blockchain networks, with the incentives scheme designed to reach equilibrium around 1,000 stake pools.

Current prominent blockchain networks are often controlled by less than 10 mining pools, exposing them to serious risk of compromise by malicious behavior – something which Cardano avoids with a system inherently designed to encourage greater decentralization.”

We’re already seeing groundbreaking projects like SingularityNET beginning to shift over to Cardano. While it may appear that Cardano has a long way to catch up with Ethereum, it’s worth remembering that – of Ethereum’s close to 3,000 apps – only a minority carry significant value or have many active users.

There is currently around $11 billion “locked up” in Ethereum’s much-vaunted DeFi projects. That’s nothing to be sniffed at, but it only takes some big projects to move to Cardano to put a big dent in Ethereum’s primary use case. It’s also worth remembering the whole space is very young with plenty of growth potential—the global legacy financial system is worth hundreds of trillions of dollars.

While multiple blockchain projects will likely co-exist, Cardano’s ability to “flip” Ethereum is looking more possible than ever.

The post Full-stack AI solution SingularityNET switches Ethereum for Cardano appeared first on AI News.

The partners are specifically looking at artificial general intelligence (AGI) technologies which aim to replicate the complex tasks of which humans are capable.

AGI is considered a step above standard AI because it aims to add human-like reasoning capabilities in uncertain scenarios. Such processing will enable AGIs to do things like solve puzzles and make plans which are beyond that of today’s AIs.

Dr Ben Goertzel, CEO of SingularityNET, said:

“The scale of the AGI deployments needed by a partner like Cisco is going to be tremendous, and we are working hard to make sure our AGI tools and our blockchain-based platform is up to the task.”

Such a large tech conglomerate as Cisco joining up with SingularityNET adds a lot of credibility to the project’s goals.

SingularityNET describes itself as the first decentralised AI network that lets anyone create, share, and monetise AI services at scale. The idea behind the decentralisation is that it prevents any single player from hoarding AI capabilities and instead for it to be distributed globally.

“To really build a global decentralized thinking machine we are going to need to put a lot of complicated ingredients together, and the collaboration between Cisco and SingularityNET has the potential to accelerate things tremendously,” Goertzel said.

In May, the Singapore and Malaysia division of Domino’s Pizza announced that it was partnering with SingularityNET to apply blockchain-powered AI to its supply chain processes.

“In 10 years, you’re going to see AGIs with IQs higher than humans,” Goertzel believes.

A beta of SingularityNET’s AI Marketplace can be found here which features 40 (as of writing) artificial intelligence-powered services.

Interested in hearing industry leaders discuss subjects like this? , , , AI &

The post Cisco partners with decentralised AI blockchain firm SingularityNET appeared first on AI News.

A report from KPMG last month explored how artificial intelligence would look to influence business. Among millennial respondents, AI came out on top as the biggest technological priority, ahead of IoT and 5G. Yet among those polled who said they were business leaders, it was less of a priority compared with robotic process automation (RPA). The report noted, however, that investing in RPA can be seen as a smoother route to future AI investment.

IT giant Infosys offers many opportunities for companies looking to invest in artificial intelligence to improve workplace productivity as well as enhance the customer experience. The company notes that customers are at different stages of their journeys depending on sector and use case; and that therefore a more nuanced approach is required.

Ahead of the AI & Big Data Expo event in Amsterdam later this month, AI News caught up with Dr. N. R. Srinivasa Raghavan, Chief Data Scientist, Data and Analytics at Infosys (left), to discuss the initiatives the company is putting together, as well as the industries set to benefit most long-term.

—

AI News: Tell us about the initiatives Infosys is putting together in the realm of artificial intelligence – and what have customers been saying?

NR: Infosys is developing several industry-oriented AI solutions, frameworks, and workbenches that can help solve business problems. Specifically, AI is being positioned for enhancing the productivity of day to day operations of our clients through automation, enable decision making at tactical and operational levels, and for better customer experience.

Our customers are at different stages of embracing AI. Depending on their needs, Infosys is able to craft solutions driven by a global team of experts in data sciences and AI, and help deploy these solutions at enterprise scale.

AI: Do you agree that many companies/vendors are using ‘AI-washing’ (much in the same way as cloud-washing years ago) and overplaying their artificial intelligence capabilities? If so – what does this mean for users and the industry at large?

NR: We believe it is not as easy to be deceptive in AI as in cloud. AI is more of an outcome heavy tech than cloud, which is more of an infrastructure play. Therefore, one can verify if indeed pattern matching and predictive science of AI is behind any software. Also, the expertise of resources working on AI projects is discernible in the quality of the output.

Nonetheless, there is possibly AI-washing in some very narrow areas like automation. It calls for better validations and governance to be put in place to weed these out.

AI: Which industries do you think are going to benefit most long-term from these kinds of technologies? How important is it that several emerging technologies – artificial intelligence, edge computing, blockchain – can all talk to each other and make each other better to provide greater business outcomes?

NR: Banking, financial services and insurance (BFSI), retail, consumer product goods and logistics (RCL), and services, utilities, resources and energy (SURE), in that order.

While AI and edge computing need to work in synchronization for cases like real time predictions, integration with blockchain will be essential in the data layers of AI. Especially where the reliability, traceability and ‘sovereignty’ of data that is feeding AI is concerned.

AI: What is the most exciting use case you have seen with artificial intelligence to date? (can be business or consumer)

NR: AI for consumer experience, AI in risk/fraud detection, AI for predictive maintenance, AI for security and AI for business functions like HR, Finance etc. These will typically be cases where there is rich and reliable data available for the AI models to work upon, as well as areas where there is executive ownership and sponsorship.

AI: What advice would you give to companies looking to embark upon or modify their digital transformation initiatives?

NR: AI is largely seen as the harbinger for an automated/human-augmented workplace. It is therefore essential to have AI experts to be part of the blue printing for the next-gen digital enterprises. AI cannot be telescoped into digital transformation programs in large enterprises. It is no more a commodity tech, but a strategic one.

AI: What are you looking forward to most at the AI & Big Data Expo and what will you be looking to tell attendees while there?

NR: We are looking forward to hearing from other participants on their success and learnings in implementing AI within Big Data context. We are eager to share our own experience in pushing the frontiers for AI and its adoption within enterprise settings.

Want to learn more about topics like this from thought leaders in the space? Find out more about the Edge Computing Expo, a brand new, innovative event and conference exploring the edge computing ecosystem.

The post Infosys: On how edge computing and blockchain will be key in different ways for AI appeared first on AI News.

For those unaware, a pump-and-dump is typically a group of individuals who use their combined buying power to artificially inflate the price of an asset before selling off near its peak. Outsiders believe the price is a sign of increased interest and end up buying high.

Anonymous messaging app Telegram is often used for organising these groups, the most notorious being ‘Official McAfee Pump Signals’ with 12,333 members.

Typically, organisers of these groups have a separate chat – sometimes accessed via a subscription fee – where users are notified ahead of those in a public group that a pump is about to occur so they can buy-in at the floor price.

Academics from Imperial College London used machine learning to predict these illicit schemes. By analysing over 300 Telegram channels, the researchers identified 220 ‘pump events’ used to build their model.

The result was the ability to predict a pump with 80 percent accuracy.

Legitimate exchanges could one day implement such an AI to warn users of an impending pump-and-dump scheme, or block them from occurring. Some exchanges, however, are involved themselves.

The researchers highlight the YoBit exchange as organising pump-and-dumps:

One case highlighted by the academics was that of the ‘BVB’ coin, a cryptocurrency said to have been created for supporters of Borussia Dortmund football club. It’s not been active since 2016.

People started buying the coin at around $0.0014 before it peaked at $0.0045, offering a potential 3x profit. The coin traded at less than its original value after just three and a half minutes.

The researchers estimate pump-and-dump schemes account for almost $7 million of monthly cryptocurrency trading volume. With a current 24 hour volume of $15.9 billion, that still only represents 0.0438 percent of the market.

You can read the full paper here (PDF)

Interested in hearing leading global brands discuss subjects like this? Find out more at the Blockchain Expo World Series with events in London, Europe, and North America.

The post AI targets cryptocurrency ‘pump-and-dump’ schemes appeared first on AI News.