Monthly Archives: August 2017

Will artificial intelligence bring a utopia of plenty? Or a dystopic hellscape? Will we, jobless and destitute, scavenge for scraps outside the walls of a few techno-trillionaires? Or will we work alongside machines, achieving new levels of productivity and fulfillment? The tech world has no lack of prognosticators: Bill Gates and Elon Musk, for example, see in AI an existential threat to the human species, while Ray Kurzweil thinks it cant come soon enough.

Silicon Slopes and big data

In fact, artificial intelligence is already here, and has been for some time. While many mistakenly equate AI with consciousnessHollywood has done the robot-gains-consciousness plot to deaththe two are distinct phenomena. As Noah Yuval Harari discusses in Homo Deus, AI need not be conscious to possess superhuman intelligence. Nor is it likely to be. Already, in domain-specific tasks, non-conscious computers are far beyond humans in intelligence. Watson beat humans at Jeopardy back in 2011; more recently, Googles AlphaGo AI beat Korean Grandmaster Lee Sodol for the fifth consecutive time at the incredibly complex game of Go. And, to those who point out the narrow scope within which such AIs can function, just remember how rapidly the scope has expanded in only a few years.

AI depends on intelligent algorithms, and such algorithms depend on the analysis of vast amounts of data. Which is why Utah is on the map with regard to AI advancement. The so-called Silicon Slopes has become, per Mark Gorenberg, a world leader in data analytics. Gorenberg should know. He serves as managing director of Zetta Venture Partners, an AI-focused venture capital firm based in San Francisco, and has invested in a number of Utah companies. The notion of analytics has become a cornerstone of Utah technology, he says.

Utah boasts high-profile data firms like Domo, Omniture (now part of Adobe) and Qualtrics, to be sure. But it also has an ecosystem of lesser-known players. Teem, for example, started by putting software on an iPad so that corporate teams could book conference rooms, Gorenberg explains. In the process, they gathered a ton of data that allows them to predict the digital workplace of the future. One Click Retail (my employerfull disclosure) uses machine learning and Amazon.com data points to help sellers optimize ecommerce operations. InsideSales employs data analytics to accelerate sales productivity by identifying the highest ROI accounts, contacts and action steps. Verscend, a healthcare analytics company, utilizes data in meaningful ways to bring our customers smarter and more effective analytics, per the company website.

But will robots dispossess us of gainful employment?

Utahs tech sector is clearly positioned to benefit from the emergence of data-driven intelligent algorithms. Well and goodbut were still left with the trillion-dollar question: Will smart machines eventually take our jobs? Are we fated to be like the typewriterthe human being as obsolete technologywhile artificial intelligence becomes, metaphorically, the word processor and laser printer? In many areas, yes. According to Gorenberg, however, there will be just as many areas in which the new AI frontier creates jobs. Sure, well lose jobs, he says. But what people arent seeing is the jobs well be gaining.

Take autonomous vehicles, Gorenberg continues by way of example. Sure, a lot of people who drive for a livingtaxi drivers, truckers, etc.will no longer be needed. At the same time, think of the downtown areas of cities. Traffic-congested urban centers no longer need be congested; sophisticated algorithms will route traffic for maximum flow. Intelligent cars, free from human errorand human distractionwill travel faster and in tighter formations, with far fewer accidents.

Then theres the issue of parking. The average downtown area uses 30 percent of its space for parking, Gorenberg notes. Those cars just sit there all day while their owners work. If the hive mind of the autonomous vehicle system knows exactly what transit is needed, and when, it can provide it at a moments notice. Fewer cars will be needed, and they can be kept outside the city center and brought in to meet demand.

Thirty percent of a citys downtown is a lot of area. Gorenberg describes the construction frenzy that will occur as the whole nature of downtowns change from all that prime acreage suddenly available. He imagines a city center could include gardens, urban manufacturing and much more.

Sure, well lose jobs. But what people arent seeing is the jobs well begaining. Mark Gorenberg, managing director, Zetta Venture Partners

And, in his vision of urban reconfiguration, Gorenberg sees beyond the myriad blue-collar jobs that such massive projects will create. Not only will you need construction workers and the like; youll need architects, city designers and planners, software development and IOT implementation, he says. There will be a need for energy experts and water experts and all of the various disciplines it takes to make a city highly functional. In short, the reuse of urban space for the next generation of cities will be a multi-trillion-dollar opportunity and will create millions of jobs at all levels.

Would you like your automation full or partial?

Economist James Bessen would agree with Gorenberg. In his article How computer automation affects occupations: Technology, jobs and skills, he concedes that full automation might indeed result in job losses. However, most automation ispartialonly some tasks are automated. In fact, as he details in his study, out of 270 occupations listed in the 1950 Census, only onethat of elevator operatorhas disappeared. Bessen claims that most job losses are not the result of machines replacing humans, but of humans using machines to replace other humans, as graphic designers with computers replaced typesetters. Or, as Mark Gorenberg puts it, this [artificial intelligence revolution] is no different than any other technology wave.

Are Bessen and Gorenberg overly optimistic, perhaps even nave, about the potential of artificial intelligence to replace humans? Or are AI alarmists a bunch of Luddites? Such questions can only be answered retrospectively. In the present, however, the incontrovertible fact is that intelligent algorithms are helping humans get better at their jobs. We dont know whether, as Alibaba CEO Jack Ma predicts, algorithms will one day be CEOs. What we do know, in the words of Gorenberg, is that a [human] CEO empowered with data is a better CEO.

So the short-to-medium-term prognosis is that human plus machine equals a better work unit than either on its own. Humans empowered by machine learning, data and sophisticated algorithms can outcompete regular old humans in the knowledge economy.

InsideSales has a dataset of over 100 billion sales interactions, says CEO Dave Elkington. The firms intelligent algorithms use this ocean of data to guide salespeople. Often, the lift provided by our software is so extreme as to make our users wonder if there might have been a reporting error. Data-powered, AI-guided salespeople. How can regular salespeople, doing things the old-fashioned way, compete? Most likely, they wont be able to.

Intelligent machines will also extend human abilities in important ways. To illustrate: the developed world (to say nothing of the developing world) faces a shortage of doctors, both generalists and specialists. I believe that AI augmenting healthcare will allow more people to perform healthcare services that today only a few can do, says Gorenberg, adding that, for example, an AI could work side by side with nurses and allow them to take expert ultrasounds and other medical images that today have to be done by a select set of experts. Thousands of high-skill nursing jobs would open up. Whats more, if lower-level professionals can do advanced medical work that is currently the exclusive domain of doctors, doctors will be free to focus on aspects of medicine for which a human with 710 years of medical training is uniquely suited.

Often, the lift provided by our software is so extreme as to make our users wonder if there might have been a reporting error. Dave Elkington, CEO, InsideSales.com

The third wave of tech revolution

If steam power was the first technological wave, and software/internet the second, artificial intelligence could well be the third. In Gorenbergs vision, the huge number of new data science and analytics positions that this upheaval will demand will compare with the millions of developer jobs created by software 25 years ago.

Over the next 510 years and beyond, well see in exactly which ways AI revolutionizes industry and business. One thing, however, is clear: Its happening, and its going to be big. And, here in Utah, were smack in the technological middle.

Jacob Andra is a writer andcontent marketing consultantin Salt Lake City, Utah. You can find him onLinkedInandTwitter.

See the rest here:

How will AI shape the workforce of the future? - Utah Business

One of the worlds most famous living artists is headed to New York City this fall, and hes bringing a massive public art project with him.

Ai Weiwei, the prolific Chinese artist and activist famously profiled in the documentary Never Sorry, is behind the ambitious Good Fences Make Good Neighbors project set to take over NYC this October. Commissioned by the Public Art Fund, the five-borough exhibition will involve over 300 locations and hundreds of individual artworks, turning the sprawling city into an unconventional canvas for his collage-like experiment.

According to a statement announcing the projects specific locations on Tuesday, Ais upcoming intervention was inspired by the international migration crisis and current global geopolitical landscape. The exhibition will use the concept of a security fence, something long touted by President Donald Trump, as its central visual element.

Ai Weiweis work is extraordinarily timely, but its not reducible to a single political gesture, Nicholas Baume, the director and chief curator of Public Art Fund, told HuffPost. The exhibition grows out of his own life and work, including his childhood experience of displacement during the Cultural Revolution, his formative years as an immigrant and student in NYC in the 1980s, and his more recent persecution as an artist-activist in China. It reflects his profound empathy with other displaced people, particularly migrants, refugees and victims of war.

The exhibition has been in development for several years, he added, so the election of President Trump has only added to its relevance.

In an earlier interview with The New York Times, Ai explained more directly that the work is a reaction to a retreat from the essential attitude of openness in American politics, though he did not explicitly mention Trumps desire to erect a wall on the border between Mexico and the U.S.

The fence has always been a tool in the vocabulary of political landscaping and evokes associations with words like border, security, and neighbor, Ai said in a statement on Tuesday. But whats important to remember is that while barriers have been used to divide us, as humans we are all the same. Some are more privileged than others, but with that privilege comes a responsibility to do more.

The name Good Fences Make Good Neighbors comes from a Robert Frost poem called Mending Wall, which Baume sent to Ai early in the projects development.The poem includes the ambiguous phrase Ai used as his title, as well as the line, Before I built a wall Id ask to know / What I was walling in or walling out / And to whom I was like to give offence.

He loved the clarity and directness of Frosts writing, and the subtle irony of this famous refrain, Baume added.

Physically, the exhibition will involve large-scale, site-specific, freestanding works, some described as sculptural interventions that will be installed in public spaces like Central Park, Washington Square Park and Flushing Meadows-Corona Park, as well as on private walls and buildings. Beyond the sculptures, Ai will display a series of 200 two-dimensional works on lamppost banners and 100 documentary images on bus shelters and newsstands. The photos were taken during the artists travels to research the international refugee crisis, and they will be coupled with text about displaced people around the world.

This is clearly not an exhibition of conventional, off-the-shelf fences, Baume said.[Ai] has taken the familiar and utilitarian material of metal fencing, which has many forms, as a basic motif. He has created multiple variations on that theme, exploring the potential of the material as a sculptural element, adapted to different locations in very site-responsive ways. Some installations are more straightforward, some more complex, but they all share this basic DNA.

Good Fences will open to the public on Oct.12 and will run until Feb. 11, marking the Public Art Funds 40th anniversary. Since its inception, the organizations mission has revolved around providing public access to contemporary art, a goal Baume said is more relevant than ever. In the past, the Fund has organized projects like Anish Kapoors Sky Mirror (2006) at the Rockefeller Center and Tatzu Nishis Discovering Columbus (2012) at Columbus Circle.

See a detailed list of the locations for Good Fences by downloading the available press releaseon Public Art Funds website. Ai Weiwei: Good Fences Make Good Neighbors will be on view from Oct. 12, 2017, to Feb. 11, 2018.

See the original post here:

Ai Weiwei Is Building Fences All Over NYC In A Powerful Public Art Project - HuffPost

Artificial intelligence labs race to develop processors that are bigger, faster, stronger.

With major companies rolling out AI chips and smaller startups nipping at their heels, theres no denying that the future of artificial intelligence is indeed already upon us. While each boasts slightly different features, theyre all striving to provide ease of use, speed, and versatility. Manufacturers are demonstrating more adaptability than ever before, and are rapidly developing new versions to meet a growing demand.

In a marketplace that promises to do nothing but grow, these four are braced for impact.

The Verge reports that Qualcomms processors account for approximately 40% of the mobile market, so their entry into the AI game is no surprise. Theyre taking a slightly different approach thoughadapt existing technology that utilizes Qualcomms strengths. Theyve developed a Neural Processing Engine, which is an SDK that allows develops to optimize apps to run different AI applications on Snapdragon 600 and 800 processors. Ultimately, this integration means greater efficiency.

Facebook has already begun using its SDK to speed up augmented reality filters within the mobile app. Qualcomms website says that it may also be used to help a devices camera recognize objects and detect object for better shot composition, as well as make on-device post-processing beautification possible. They also promise more capabilities via the virtual voice assistant, and assure users of the broad market applications--from healthcare to security, on myriad mobile and embedded devices, they write. They also boast superior malware protection.

It allows you to choose your core of choice relative to the power performance profile you want for your user, said Gary Brotman, Qualcomm head of AI and machine learning.

Qualcomms SDK works with popular AI frameworks, including Tensor Flow, Caffe, and Caffe2.

Googles AI chip showed up relatively early to the AI game, disrupting what had been a pretty singular marketplace. And Googles got no plans to sell the processor, instead distributing it via a new cloud service from which anyone can build and operate software via the internet that utilizes hundreds of processors packed into Google data centers, reports Wired.

The chip, called TPU 2.0 or Cloud TPU, is a followup to the initial processor that brought Googles AI services to fruition, though it can be used to train neural networks and not just run them like its predecessor. Developers need to learn a different way of building neural networks since it is designed for Tensorflow, but they expectgiven that the chips affordabilitythat users will comply. Google has mentioned that researchers who share their research with the greater public will receive access for free.

Jeff Dean, who leads the AI lab Google Brain, says that the chip was needed to train with greater efficiency. It can handle180 trillion floating point operations per second. Several chips connect to form a pod, that offers 11,500 teraflops of computing power, which means that it takes only six hours to train 32 CPU boards on a portion of a podpreviously, it took a full day.

Intel offers an AI chip via the Movidius Neural Compute Stick, which is a USB 3.0 device with a specialized vision processing unit. Its meant to complement the Xeon and Xeon Phi, and costs only $79.

While it is optimized for vision applications, Intel says that it can handle a variety of DNN applications. They write, Designed for product developers, researchers and makers, the Movidius Neural Compute Stick aims to reduce barriers to developing, tuning and deploying AI applications by delivering dedicated high-performance deep-neural network processing in a small form factor.

The stick is powered by a VPU like what you might find in smart security cameras, AI drones, and industrial equipment. It can be used with trained Caffe framework-based feed-forward Convolutional Neural Network or the user may choose another pre-trained network, Intel reports. The Movidius Neural Compute Stick supports Cnn profiling, prototyping, and tuningworkflow,provides power and data over a single USB Type A port, does not require cloud connectivity, and runs multiple devices on the same platform.

From Raspberry Pi to PC, the Movidius Neural Compute Stick can be used with any USB 3.0 platform.

NVIDIA was the first to get really serious about AI, but theyre even more serious now. Their new chipthe Tesla V100is a data center GPU. Reportedly, it made enough of a stir that itcaused NVIDIA's shares to jump 17.8% on the day following the announcement.

The chip stands apart in training, which typically requires multiplying matrices of data a single number at a time. Instead, the Volta GPU architecture multiplies rows and columns at once, which speeds up the AI training process.

With 640 Tensor Cores,Volta is five times faster than Pascal and reduces the training time from 18 hours to 7.4 and uses next generation high-speed interconnect technology which, according to the website, enables more advanced model and data parallel approaches for strong scaling to achieve the absolute highest application performance.

Heard of more AI chips coming down the pipe? Let us know in the comments below!

See the article here:

Intel, Qualcomm, Google, and NVIDIA Race to Develop AI Chips and Platforms - All About Circuits