Daily Archives: July 18, 2021

All the sessions from Transform 2021 are available on-demand now. Watch now.

Technology industry veterans Tom Davenport and Tom Seibel have seen firsthand how data, analytics and artificial intelligence have changed business models over the past three decades.

In a conversation with ThoughtSpot Chief Data Strategy Officer Cindi Howson at VentureBeats Transform 2021 conference on Tuesday, Davenport and Siebel shared their insights into how technology has boosted innovation and how it can transform industries, as well as the dangers it presents.

Davenport said the biggest change hes seen in his career has been the democratization of technology.

Theres been a continual move toward the software being easier to use, and being able to do more things on its own automated analytics, automated data science, automated machine learning, Davenport said. I think were poised for even more democratization, which is great. I think overall theres some issues that it raises, but it really opens up this field to a lot more people who may not have been nerdy enough to study statistics and get into the details of how you create various models.

Siebel said its difficult to overestimate the impact of the cloud.

Computing used to be expensive. Storage is something we used to move in and out of machinery with a forklift, he said.

Now, infinite computing capacity and storage are essentially free. And that has allowed startups to grow quickly and compete with more established firms.

They dont have to build a big IT infrastructure, they can get it pretty much all from the cloud, Siebel said.

Technologies like cloud computing and predictive analytics have enabled classes of applications that solve problems that were previously unsolvable, Siebel said.

We have large banks, the oil and gas companies, Department of Defense, Food and Drug Administration, travel, transportation companies, all kind of reinventing themselves using predictive analytics to fundamentally change the way that they manage their businesses and deal with customers, Siebel said.

Despite its benefits, AI has some pitfalls, Howson noted. In 2020, for instance, some felt like their models did not help predict the future.

Davenport attributed this to the slow pace in which humans and organizational culture improve.

Particularly in these legacy companies, we havent made much progress in becoming more data-driven in how we make decisions, Davenport said. Theres still a lot in organizations, conservatism, failure to experiment with new technologies and new approaches, to analyzing and understanding data. So that to me is the big problem.

Additionally, AI presents some ethical issues.

Siebel, who believes the largest commercial application of AI will be in precision medicine, said we can build machine learning models that can do meaningful things like disease prediction. But health care providers via the state or private enterprise can misuse that data, he said.

Are they going to use it to ration health care? Absolutely, Siebel said. Are they going to use it to set rates? Absolutely. This is very scary stuff.

He described the privacy and ethical implications of AI very troubling.

They need to be addressed, Siebel said.

Read more:

Trilogy of Data, analytics, AI is accelerating innovation across industries - VentureBeat

The new AI system is inspired by humans. When humans see the color of one object, they can easily apply it to other objects by replacing the original color with a new color. Credit: Chris Kim

A team of USC researchers is helping AI imagine what is invisible. This approach could also lead to fairer AI, new drugs, and improved safety for self-driving cars.

Imagine an orange cat. Now imagine the same cat. However, it has black fur. Now, imagine a cat pretending to be along the Great Wall of China. When you do this, the activation of a series of neurons in the brain brings to mind variations of the images presented, based on previous world knowledge.

In other words, as a human, its easy to imagine objects with different attributes. However, despite advances in deep neural networks that rival or exceed human performance on certain tasks, computers still struggle with the very human skill of imagination.

Currently, a USC research team consisting of Professor Laurent Itti of Computer Science and doctoral students Yunhao Ge, Sami Abu-El-Haija, and Gan Xin has used human-like features to date. We have developed an AI that imagines no different objects. attribute.Title treatise Zero-shot synthesis by learning with group supervised learningWas published in the 2021 International Conference on Learning Representation, held on May 7th.

We were inspired by human visual generalization, which attempts to simulate human imagination with machines, said Ge, the lead author of the study.

Human can separate learned knowledge by attributes (shape, pose, position, color, etc.) and recombine them to imagine new objects. Our paper uses neural networks. Im trying to simulate this process.

For example, suppose you want to create an AI system that produces an image of a car. Ideally, the algorithm would provide some images of the car, allowing it to generate different types of cars of any color from different angles, from Porsche to Pontiac to pickup trucks.

This is one of AIs long-standing goals of creating extrapolable models. This means that, given some examples, the model should be able to extract the underlying rules and apply them to a wide range of new examples that we have never seen before. However, machines are most commonly trained on sample features, such as pixels, without considering the attributes of the object.

In this new study, researchers are trying to overcome this limitation by using a concept called unraveling. Unraveling can be used to generate deepfake, for example, by unraveling human facial movements and identities. By doing this, people can synthesize new images and videos that replace the identity of the original person with another person, but maintain the original movement, Ge said.

Similarly, the new approach takes a group of sample images instead of one sample at a time as in traditional algorithms, mines the similarities between them, and controllably entangled expression learning. To realize what is called.

We then recombine this knowledge to achieve what we call new controllable image composition, or imagination. For example, in the Transformers movie, you can take the shape of a Megatron car, the color and pose of a yellow Bumblebee car, and the background of Times Square in New York. As a result, this sample during a training session. Even if he wasnt witnessed, a Bumblebee-colored Megatron car would be driven in Times Square.

This is similar to how humans extrapolate. When humans see the color of one object, they can easily apply it to other objects by replacing the original color with a new color. Using their technology, the group has generated a new dataset containing 1.56 million images that may be useful for future research in this area.

Unraveling is not a new idea, but researchers say their framework is compatible with almost all types of data and knowledge. This opens up new application opportunities. For example, create a fairer AI by unraveling race- and gender-related knowledge and completely removing sensitive attributes from equations.

In the field of medicine, unraveling the functions of medicine from other properties and recombining them to synthesize new medicine may help doctors and biologists discover more useful drugs. You can also create a safer AI by infusing your machine with imagination, for example, by allowing self-driving cars to imagine and avoid dangerous scenarios never seen before during training.

Deep learning has already demonstrated outstanding performance and expectations in many areas, but often this is done by shallow imitation and does not have a deep understanding of the individual attributes that make each object unique. Said Itti. For the first time, this new unraveling approach unleashes the new imagination of AI systems and brings it closer to understanding the human world.

Reference: May 7, 2021, Zero Shot Synthesis by Group Supervised Learning by Yunhao Ge, Sami Abu-El-Haija, Gan Xin, Laurent Itti 2021 International Conference on Learning Representation..Link

Researchers Enable AI To Use Its Imagination Closer to Humans Understanding of the World Source link Researchers Enable AI To Use Its Imagination Closer to Humans Understanding of the World

See the original post:

Researchers Enable AI To Use Its Imagination Closer to Humans' Understanding of the World - California News Times

A documentary on the late chef Anthony Bourdain showcases the extraordinary advances of artificial intelligence (AI). Roadrunner uses Deep Fake technology to show Bourdain mouthing words he never uttered. Its a wake-up call on potential side effects of AI. Deep Fakes are a subfield of AI that allow realistic forgeries of both video and audio. The speed of advances in AI have made it possible to create Deep Fakes using freely available software and computer processing power that can be rented. AI is perhaps the most transformative technology under development. Consequently, it also brings about entirely new risks.

Deep Fakes pose a fundamental danger. They can quickly undermine trust, the invisible bond that holds many collectives together. There are many examples of Deep Fakes being used by state actors to influence elections and sow seeds of discord. The US has been a victim of Deep Fakes. Even in India, Deep Fakes are known to have been circulated in some electoral contests. AIs rapid upgrades, coupled with a transition to digital modes of governance, calls for a new level of safeguards on the part of GoI.

One example is the way the US is gearing up to face new challenges. Its been hit repeatedly by ransomware attacks, where hackers introduce malicious software code into networks to prevent victims from accessing their data. This year an oil pipeline operator in the US was hit by ransomware and was forced to pay the hackers. American responses are no longer reactive. The US is investing in talent and technology needed to cripple hackers cyber infrastructure. Such enormous challenges mean GoI must build on its existing technology talent pool. Penny-pinching and red tape shouldnt come in the way. Social and economic costs of unchecked Deep Fakes and ransomware will be far greater.

This piece appeared as an editorial opinion in the print edition of The Times of India.

END OF ARTICLE

Read the original post:

AIs dark side: Deep Fakes are easy to create. GoI needs to invest in tech & talent to counter the new dange - The Times of India Blog

[author: Doug Austin, Editor of eDiscovery Today]

The final installment of the virtual Legalweek(year) just concluded, with the fifth installment having completed this week, after previous iterations in February (the traditional time of year for in-person Legalweek), March, April, and May. I attended a handful of sessions and planned to give you a sampling of the sessions I attended, but the session that I attended at the end of the conference on Wednesday was so good, I decided to cover it specifically.

The session was The Ethics of AI in The Legal Profession and it was conducted by Tess Blair of Morgan Lewis and Maura R. Grossman of the University of Waterloo and Maura Grossman Law (who should be a familiar name to any of you who understand Technology Assisted Review (TAR) as she and Gordon V. Cormack defined the term and issued the groundbreaking study that demonstrated how TAR could be more efficient and effective for document review).

Blair and Grossman covered several aspects of the use of AI, a couple of which I will briefly recap here. They also provided some interesting graphics to illustrate various concepts such as machine learning (interspersed pictures of chihuahuas and blueberry muffins so similar its startling), natural language processing (NLP) and deep learning.

Advising Clients Developing or Using AI

Among the topics covered here were the idea of how crowdsourcing can introduce bias into AI algorithms, where the example used was to type in the phrase lawyers are into Google search and the completion terms that dropped down included terms like scum, liars, sharks, evil, and crooks. Ouch!

There are three places where AI bias can come into play: 1) the data, the example used was to use only white faces to train an algorithm to the point it wont handle black faces; 2) the algorithm, which can be tuned to weight things differently; and 3) humans, which may have an algorithm aversion, may have automation bias (where the algorithm is assumed to be correct because it is not human) or may have confirmation bias (where they agree only if the results confirm what they already believe).

Grossman also discussed the use of the Correctional Offender Management Profiling for Alternative Sanctions (COMPAS), which tended to rate the possibility for black defendants to reoffend at much higher score than non-black defendants, as was the case of an 18 year old black woman who was rated high risk (a score of 8) for future crime after she and a friend took a kids bike and scooter that were sitting outside whereas a 41 year old white man who had already been convicted of armed robbery and attempted armed robbery was rated a low risk (according to this article by ProPublica).

Grossman pointed out that COMPAS experienced function creep where it was originally designed to provide insight into the types of treatment an offender might need (e.g., drug or mental health treatment), then expanded to decision making about conditions of release after arrest (e.g., release with no bail, bail, or retention without bail), before being expanded again to decisions about sentencing.

Blair added discussions regarding privacy and also AI moral dilemmas, such as the case of an autonomous vehicle, entering a tunnel with child in the middle of the road and a decision to make whether to go straight ahead and kill the child or veer off into the wall and kill the passenger (yikes!). Those and other moral dilemmas can be found here at MITs Moral Machine site.

Resources for Practicing with AI

With regard to practicing law while using AI, the presenters discussed several sources of guidance with regard to an attorneys duty for understanding AI, including:

Conclusion

Blair and Grossman concluded with a brief discussion on whether AI is going to take lawyer jobs (there may be fewer attorneys in the future, but they will be more focused on the types of tasks they were trained for in law school) and whether AI will ever become smarter than humans (Grossman stated that the technology often still cant do what a 3-year-old can do).

So, with great power comes great responsibility.

But lawyers (or anybody using AI) need to do their part to understand the technology and the concepts (as well as the risks) to fully benefit from AI. When they do, they can accomplish amazing things!

Next year, Legalweek returns to an in-person event in New York City from January 31 from February 3! See you there!

[View source.]

The rest is here:

The Ethics of AI in The Legal Profession | Ipro Tech - JDSupra - JD Supra

As more fast food restaurants continue to experiment with plant-based protein alternatives - Little Caesars 'Planteroni' Pizza in a partnership with Field Roast being one of the most recent examples plenty of consumer still havent bought into the fake meat craze.

Part of the problem behind consumers stubborn resistance to adopting more of a plant-based or flexitarian diet is the common, and fair, complaint that many of these meat alternatives taste off. Soy-based patties can leave an unpleasant aftertaste, and never mind the crumbly texture. Beyond Meats fake chicken of yesteryear offered notes of carrots, frozen peas and fava beans, and not necessarily in a good way. And while Beyonds and Impossible Foods plant-based protein substitutions for burgers are about the closest thing one can get to the real thing, the coconut oil can result in traces of sweetness that can be off-putting. Of course, all of these are improvements over the veggie burgers from many years ago, which would have tasted far better if they were allowed to remain as vegetables.

But what if flavorings could help make these new and futureplant-based proteinproducts more palatable to more consumers? After all, the companies that are driving the multi-billion dollar global flavor and fragrance industry keep developing ingredients that are increasingly more sophisticated and are in just about in every product we put in or on us. Many of us are already taking such action in our kitchens vanilla extract, for example, is a common ingredient in our cupboards because it easily binds to proteins without giving off its flavoring and it masks other flavors that could otherwise taste unpleasant.

Now, lets add another challenge, and one that if overcome, could help plant-based protein scale up: which of course would help wean more consumers away from the carbon- and water-intensive meat and dairy industry.

One hurdle that food companies face is thatdevelopingnew products that will eventually be accepted by the masses can take years. But what if that process could be shortened by harnessing the potential of artificial intelligence (AI)?

To that end, Firmenich, the Switzerland-based fragrance and flavoring giant, says its on a path toward improving the taste of plant-based protein products.

The company recently announced the launch of what it says is the first flavor developed by AI, a lightly grilled beef taste that could be used in plant-based foods. Emphasis should be put on lightly grilled, as yet another complaint of fake meat is that no matter how its cooked, it can often leave dinerswith a cringeworthy charred taste.

Firmenich is understandably mum on what flavor notes are exactly in this new AI-induced flavor profile. Nevertheless, the company could provide that final piece of the puzzle to companies that seek to recreate the flavor and texture of meat: and not 95 percent, not 99 percent, but a 100 percent success in copying meats flavor profile. So far, companies like Beyond and Impossible Foods have pretty much nailed the texture part. What has proven to be difficult is finding that complexity behind the flavor of meat, which includes countless factors such as umami, fat and of course, how it was cooked, such as on a flame or in an oven. And, as we all know if we dont follow directions, the way in which we cook our foods can affect the flavor: so, these plant-based foods need to hold their flavor profile as much as possible whether they are grilled, baked, toasted or out of desperation (ew!) even microwaved.

A company like Firmenich has the advantage of access to a wide range of flavorings within its labs. What it does not have is the time to test out the infinite number of flavor combinations. Therein lies the power of AI the company recently described its database of flavorings to Phys.org as a piano with 5,000 keys that through algorithms allows its staff to test out different combinations. The results could include even more plant-based options coming soon to a supermarket near you.

Image credit: Rolande PG/Unsplash

Follow this link:

How AI Could Improve the Taste of Plant-Based Protein - Triple Pundit

Cloud contact center provider Five9 this week announcedthat it has revamped its intelligent virtual agent (IVA) development platform, added pre-built IVA applications for the health vertical, and tapped partners for integration of voice biometrics, real-time speech analytics, and agent coaching.

These updates reflect the desire Five9 sees among its customers to increase their use of conversational self-service along with live agent support, Callan Schebella, EVP of product management at Five9, shared with No Jitter via email. Five9 sees enormous demand for IVAs as customers seek to increase automation rates to manage costs while improving the customer experience, he added.

IVA Development, Tasks, Integrations

To improve performance and enable development of new capabilities, Five9 has rearchitected and rewritten the underlying layers of Inference Studio, its no-code IVA development platform, andalso provides access to a wide range of conversational technologies from leading companies like Amazon Web Services, Google Cloud, IBM, and others, Schebella said.

That rearchitecting work included creating a media server to support VoiceXML and enable call controls, for delivery of new features like call recording, passive voice biometrics, and more, he added.

Using Studios visual, browser-based interface, developers can design IVA applications by dragging and dropping nodes onto a canvas to build a task flow, Schebella said. As part of the update, Five9 has optimized the interfaces task flow editor for larger canvases that can support more nodes thus reducing the load time when building task flows. Additionally, it has applied reverse indexing to datastores, for faster loading of data, he said.

Other Studio platform enhancements include a new user interface design, a customized development process for messaging applications, and improved reporting and maintenance of IVA tasks and call flows via new dashboard, as shown below. The Studio dashboard is similar to a contact center wall board, where you can monitor IVA usage, call volumes, and chat volumes in real time, Schebella said.

Studio simplifies IVA development even further by providing access to a Task Library, which functions similarly to an app store and contains over 40 pre-built IVA application templates from Five9 and partners that organizations can use as blueprints for their own customized IVAs. With this weeks announcement, Five9 has added its first set of vertical-specific tasks to the library, for healthcare and health insurance providers. Tasks for other verticals, such as retail and financial services, will follow, Five9 said.

The health industry vertical suite includes tasks for appointment scheduling, health plan enrollment, FAQ, prescription management, and test results. To describe how an organization might customize one of these tasks, Schebella shared how it might add its own greeting, prompts, and set of frequently asked questions and answers to the FAQ templates pre-configured task flow. It might also choose preferred text-to-speech voices and languages, he added.

Voice AI Partners

Join Five9 this fall for Enterprise Connect 2021, and catch up on the latest trends and technologies for communications and collaboration, including CCaaS platforms. As a No Jitter reader, use the code NJAL200 to save $200 on your registration.

See the original post:

Five9 Brushes Up on AI in CCaaS Platform - No Jitter

Leveraging artificial intelligence (AI) provides companies with a unique and enduring competitive advantage, witnessed by the fact that AI-first companies are the worlds only trillion-dollar companies. Thats the view of Ash Fontana, a leading global start-up investor with a specialist interest in artificial intelligence and author of The AI First, How to Compete and Win with Artificial Intelligence.

AI is the one that compounds most quickly and is the hardest to catch up to. Once you build it, it becomes a loop and it builds itself which is why it is so powerful, he tells The Irish Times.

In the book, Fontana frames AI as the third wave of economic development.

The first wave, the physical, dates back to the Stone Age. Think rope traps and spears, tools that allowed us to go beyond our immediate physical reach to gather more food than we could with our bare hands. However, this physical leverage was limited by scale and our intellectual capacity.

The second wave brought intellectual leverage. It started with the printing press which allowed us to distribute information, with computers following, extending our intellectual reach. Insights were limited, however. Artificial intelligence, the third wave, provides decision-making leverage, he says.

Making fabrics

Fontana uses the example of producing fabrics to illustrate the point. The physical age brought the loom which made it possible to stitch fabrics faster than by hand. The information age allowed computers to turn drawings into patterns for the loom to weave.

The third wave changes the game: computers scan photos on social media, figure out consumer trends, draw up new styles and turn drawings into patterns for the loom. New styles hit the stores just as they become fashionable.

In the world of AI, the collection of data is merely the starting point. Deploy the right network of interlocking datasets, filters and tools and you can develop a flywheel effect.

Consider the vast amount of money Google invests in its Google Maps for instance or Amazon on its Alexa-based models, both of which hoover up huge amounts of useful information. As Fontana notes, these are not stand-alone products for either company but part of a suite of products that helps fulfil their data strategy.

Put simply, the power of AI lies in its capacity to turn data into really useful information to aid decision-making. Big tech does it at scale with sometimes alarming efficiency, but Fontana says small businesses can use it too. A sandwich shop can use simple AI tools and techniques to monitor its inventory on the shelves during the day and recalibrate its prices.

In his book he also explains how a lean version of AI can be employed by almost any business.

Many businesses have AI working in the background, whether they are conscious of it or not. Payment service providers such as banks and credit card companies employ it to detect fraud, for example.

Consider the example of point-of-sale solutions group Square. It has the capacity to access real-time information from the tills of its retail customers with their permission which can be used to inform decisions made by its lending subsidiary, Square Capital. By comparing the data from one retail customer to a range of till information from other similar customers it has already lent to, it can predict whether a customer is good for a loan.

You dont get a loan if you dont add data and you qualify for a loan because your data can be compared with other data to make a prediction.

Success in becoming an AI-first company, he says, is about how effectively you can collect the right information and use it to create good predictive models. Master this and you create powerful network effects to outpace your competitors.

First organise the data you have. Spend the time and money you need to get everyone storing data in the right place, with all the tags needed to show context around any particular dimension of the data. Then try some of the more basic machine learning methods available in free easy-to use software packages, he advises.

Fontana distinguishes between entry level and what he calls next level network effects. A simple entry level effect, for example, might tell you that half of your customers are women over 45.

The next level is when the machine is automatically learning over lots of data points throughout a network and is then also creating predictive information We think the next person to buy your product will have the following attributes ...

The most common mistake when trying to become an AI-first company is not having everyone aligned with an AI strategy, he says, in other words, not thinking about where to get data, how to process it into information and build models that generate data network effects.

The next most common mistake is not investing enough in security, infrastructure and governance.

AI is already being deployed as a management tool in traditional offices and work-from-home settings as a way of measuring productivity and possibly engagement, with obvious concerns on the part of some employees.

It is also being using used as an active listening tool. For example, calls to a customer service centre can be monitored to determine customer pain points. Insights from this information can then be incorporated into the design of new products and services.

Fontana says the sci-fi nightmare of rogue robots and supercomputers remains very much in the realm of fiction. We are not at the point where machines are doing things that humans are not ultimately controlling.

AI can be weaponised, but I think ultimately thats about how humans behave and not AI doing something by itself.

Ethical concerns about surveillance capitalism by big tech abound but Fontana argues that AI can be a force for wider good. Consider its role in combatting the Covid 19-pandemic, he says. The development of vaccines and the choice of the most efficacious drugs to treat Covid-infected patients was accelerated through AI while the complex logistical challenges of the mass rollout of vaccines was also helped by the use of AI.

Panel: Principles of Lean AI

Distinguish Lean AI from Lean start-ups: In a start-up, you are looking to develop a minimum viable product. In Lean AI, you are attempting to make an existing model more accurate. The output is a prediction, not a calculation.

Make better predictions: The aim is to get to a point where the prediction starts getting better than a humans. A prediction often takes the form of a classification, for example classifying the information available in a photograph.

Be selective: Collecting data and performing calculations can be expensive. Spend time figuring out exactly what you want to predict in order to settle on the prediction usability threshold, the point at which a prediction becomes useful.

Play with statistics: Get one answer with one statistical method then use that to discover the next answer using another statistical method.

Be clear on your aim: Lean AI is about solving a specific problem with AI and building a small but complete AI-first product that can either grow into other domains or remain focused on one.

The AI-First Company, how to compete and win with artificial intelligence, by Ash Fontana is published by Penguin/Portfolio

Read this article:

How companies can use AI to get ahead of the competition - The Irish Times