At SC19: Who’s Accountable ‘When Technology Kills’?

By Doug Black

December 2, 2019

“One of the things we are witnessing is the compute requirement for (AI) training jobs is doubling every three-and-a-half months. So we were very impressed with Moore’s Law doubling every 18 months, right? This thing is doubling every three and a half months. Obviously, it’s unsustainable. If we keep at that rate for sustained periods of time, we will consume every piece of energy the world has just to do this.”

– Dario Gil, director, IBM Research

The relationship between humans and machines is inarguably one of the great issues of this century. Machine intelligence is rising at supersonic speed (see above); human intelligence is, at best, rising slowly. Machines, though bereft of EQ, can apply unimaginably, inhumanly high (and getting higher) IQ to its assigned tasks; human EQ can be nurtured, but our ability to comprehend and synthesize large amounts complex information has limits. Machines are capable of handling larger spheres of our work and personal lives; people are ceding more personal and work tasks to machines. These trend lines are firmly in place and – as things stand now – will only accelerate.

“When Technology Kills,” the conference plenary at last week’s SC19 in Denver, comprised of three technology intellectuals, took up thorny AI-related questions around how much autonomy ML, AI, IoT and smart data should be given, regulations for governing software development and data privacy, additional training people should receive to manage new software systems, and ultimately, who is responsible when software fails, violates our legal rights, causes property damage, injury or loss of life?

One would assume these questions weigh heavily on the minds of the HPC-AI community, many of whose most capable and influential technologists were gathered at SC19. But the plenary session attracted a relatively light turnout – tending to confirm the perception that the AI industry is far more focused on developing systems with superhuman capabilities than on the social and ethical implications those systems pose.

To be fair, technologists above all are tasked with building powerful technology that disrupts and drives their organizations to first mover status. It’s also true that philosophical discussions of AI ethics can be somewhat ethereal, light on actionable insight. But a growing chorus of voices are being raised in industry analyst, academic, political and journalistic circles about controlling the ways in which humans and machines will co-exist in the decades to come. Their warning to the technology industry: Avoid issues of AI ethics and accountability at your own peril.

In fact, the common current running through the plenary discussion centered more on people than machines, and on the need to balance the legitimate though often conflicting interests of technology’s constituent groups.

SC19 Plenary panel, from left: Keri Savoca, freelance technical writer; Eric Hunter, Bradford & Barthel; Erin Kenneally, Elchemy; Ben Rothke, Tapad

As Eric Hunter, futurist and director of knowledge, innovation and technology strategies at consulting firm Bradford & Barthel LLP, said, “Oftentimes, people will focus on the technology itself. (But) it really comes down to the individuals involved, the humans involved. That can sound profoundly redundant, but I’m saying that because you can’t divorce human behavior and technology. And it’s about the individuals that are adapting to these technologies, creating them, interacting with them. And then when something fails, what was the decisions that were made? And what can be learned from the individuals involved?”

Early in the plenary, Erin Kenneally, CEO of advisory group Elchemy, offered a coda for the session:

“I think, when you use the terms responsibility and blame, implicit in that is this notion of lack of trust. And I think one of the major problems that we’re facing now is this gap between our technology and our laws. I (call it) a gap between our expectations and our capabilities… Imagine a graph and it’s… got two lines and the upper line is very steeply sloping, approximating Moore’s Law. And that represents the rate of change of technology capabilities… At the bottom, you’ve got, not quite flatlining but slightly up-slope is a line that represents our laws, and that is our expectations. So there’s this gap … and this is where we have conflicts of rights and interests.“Let’s take deploying AI edge devices to monitor your behavior. The issue is my rights. My rights and my interest in privacy may conflict with my employer’s right and interest in the security of their enterprise or their commercial free speech rights. And that may conflict with my fellow citizens’ interest in their own security and privacy, which may conflict with the government’s interests in securing critical infrastructure. We see these instances all over the place.

“Like I mentioned with AI, we’ve got recommender systems, scoring systems, classification systems that are all trying to build predictive models about us. At the end of the day, I think it’s important to realize that technology is no longer just providing affordances. It’s not just spell checking our documents. It’s actually making decisions and taking actions by and for and with us, and those impact our rights and interests. And oftentimes, those decisions and those actions are being done in a very asymmetrical, opaque manner. And they have impact and we’re not certain of those impacts, sometimes. So we’re dealing with this widening delta between our capabilities and our expectations. And I think that’s what we have to worry about.

“I think there are at least three consequences…, you get increasing tensions between legitimate stakeholders. It’s easy to say good guy versus bad guy we know someone’s right and someone’s in the wrong, but when you’ve got good guy versus good guy versus good guy, how do you resolve those issues? You’ve also got an inefficient avoidance of risk problems as well if organizations don’t know if what they’re doing is violating the law. And then finally, you’ve got an undermining of ordering forces, people either receding from the marketplace and not trusting technology, or going rogue and extrajudicial and taking the law into their own hands. And we don’t, we don’t want any of those situations to happen.

“There are at least five ‘trust mechanisms’ that we can rely on.

“Number one, we need incentives to build secure software and not just race to time-to-market pressures. We need to rely on responsible research and development. I think that’s critical.

“Secondly, we need to do a better job of getting real-world, longitudinal, large scale data in front of researchers and developers to test and evaluate their technologies. …You can build the greatest algorithm since sliced cheese but if you don’t have good data going in you’re going to get bad results.

“We need to do a better job from a smart governance perspective with regard to the collection, use and disclosure of data, and we need to be more innovative from a legal perspective with regard to liability and holding people responsible.

“We need to have better convening and coordinating between academia, research and the private sector around standards and best practices.

And then finally, I think ethics underpins all of those… I like to provide people with ‘framing thoughts.’ If you think of ethics, think of it in terms of the three legged stool: principles, application of those principles and then enforcement. We’re doing okay on the first leg, a little bit better on the second and we need to do a lot of work on the third.”

We’ll end this brief review of the hour-long discussion with a comment from Ben Rothke, author and senior information security specialist at marketing consulting firm Tapad, who placed matters of ethics and security squarely at the feet of senior vendor management.

“When you look at it from a corporate governance perspective, when a large organization has record profits, we see senior management often needs to reap the (financial) benefits there,” he said. “Then it needs to trickle down also that they’re responsible for the safety, the information security. And often the CEO in the aviation sector, it’s a matter of does his organization have a safety culture? Where there’s a lot of accidents with a lot of incidents it’s because management didn’t develop a safety culture. And so too that works in information, security, privacy and everything. If management takes it seriously, if it’s an imperative to them, then it will trickle down. So really, at the end of the day, everything starts at the top, you have to build that culture. And where that exists, it will trickle down. If they take it seriously, the rest of the organization will take it seriously.”

Subscribe to HPCwire's Weekly Update!

Be the most informed person in the room! Stay ahead of the tech trends with industry updates delivered to you every week!

MLPerf Inference 4.0 Results Showcase GenAI; Nvidia Still Dominates

March 28, 2024

There were no startling surprises in the latest MLPerf Inference benchmark (4.0) results released yesterday. Two new workloads — Llama 2 and Stable Diffusion XL — were added to the benchmark suite as MLPerf continues Read more…

Q&A with Nvidia’s Chief of DGX Systems on the DGX-GB200 Rack-scale System

March 27, 2024

Pictures of Nvidia's new flagship mega-server, the DGX GB200, on the GTC show floor got favorable reactions on social media for the sheer amount of computing power it brings to artificial intelligence.  Nvidia's DGX Read more…

Call for Participation in Workshop on Potential NSF CISE Quantum Initiative

March 26, 2024

Editor’s Note: Next month there will be a workshop to discuss what a quantum initiative led by NSF’s Computer, Information Science and Engineering (CISE) directorate could entail. The details are posted below in a Ca Read more…

Waseda U. Researchers Reports New Quantum Algorithm for Speeding Optimization

March 25, 2024

Optimization problems cover a wide range of applications and are often cited as good candidates for quantum computing. However, the execution time for constrained combinatorial optimization applications on quantum device Read more…

NVLink: Faster Interconnects and Switches to Help Relieve Data Bottlenecks

March 25, 2024

Nvidia’s new Blackwell architecture may have stolen the show this week at the GPU Technology Conference in San Jose, California. But an emerging bottleneck at the network layer threatens to make bigger and brawnier pro Read more…

Who is David Blackwell?

March 22, 2024

During GTC24, co-founder and president of NVIDIA Jensen Huang unveiled the Blackwell GPU. This GPU itself is heavily optimized for AI work, boasting 192GB of HBM3E memory as well as the the ability to train 1 trillion pa Read more…

MLPerf Inference 4.0 Results Showcase GenAI; Nvidia Still Dominates

March 28, 2024

There were no startling surprises in the latest MLPerf Inference benchmark (4.0) results released yesterday. Two new workloads — Llama 2 and Stable Diffusion Read more…

Q&A with Nvidia’s Chief of DGX Systems on the DGX-GB200 Rack-scale System

March 27, 2024

Pictures of Nvidia's new flagship mega-server, the DGX GB200, on the GTC show floor got favorable reactions on social media for the sheer amount of computing po Read more…

NVLink: Faster Interconnects and Switches to Help Relieve Data Bottlenecks

March 25, 2024

Nvidia’s new Blackwell architecture may have stolen the show this week at the GPU Technology Conference in San Jose, California. But an emerging bottleneck at Read more…

Who is David Blackwell?

March 22, 2024

During GTC24, co-founder and president of NVIDIA Jensen Huang unveiled the Blackwell GPU. This GPU itself is heavily optimized for AI work, boasting 192GB of HB Read more…

Nvidia Looks to Accelerate GenAI Adoption with NIM

March 19, 2024

Today at the GPU Technology Conference, Nvidia launched a new offering aimed at helping customers quickly deploy their generative AI applications in a secure, s Read more…

The Generative AI Future Is Now, Nvidia’s Huang Says

March 19, 2024

We are in the early days of a transformative shift in how business gets done thanks to the advent of generative AI, according to Nvidia CEO and cofounder Jensen Read more…

Nvidia’s New Blackwell GPU Can Train AI Models with Trillions of Parameters

March 18, 2024

Nvidia's latest and fastest GPU, codenamed Blackwell, is here and will underpin the company's AI plans this year. The chip offers performance improvements from Read more…

Nvidia Showcases Quantum Cloud, Expanding Quantum Portfolio at GTC24

March 18, 2024

Nvidia’s barrage of quantum news at GTC24 this week includes new products, signature collaborations, and a new Nvidia Quantum Cloud for quantum developers. Wh Read more…

Alibaba Shuts Down its Quantum Computing Effort

November 30, 2023

In case you missed it, China’s e-commerce giant Alibaba has shut down its quantum computing research effort. It’s not entirely clear what drove the change. Read more…

Nvidia H100: Are 550,000 GPUs Enough for This Year?

August 17, 2023

The GPU Squeeze continues to place a premium on Nvidia H100 GPUs. In a recent Financial Times article, Nvidia reports that it expects to ship 550,000 of its lat Read more…

Shutterstock 1285747942

AMD’s Horsepower-packed MI300X GPU Beats Nvidia’s Upcoming H200

December 7, 2023

AMD and Nvidia are locked in an AI performance battle – much like the gaming GPU performance clash the companies have waged for decades. AMD has claimed it Read more…

DoD Takes a Long View of Quantum Computing

December 19, 2023

Given the large sums tied to expensive weapon systems – think $100-million-plus per F-35 fighter – it’s easy to forget the U.S. Department of Defense is a Read more…

Synopsys Eats Ansys: Does HPC Get Indigestion?

February 8, 2024

Recently, it was announced that Synopsys is buying HPC tool developer Ansys. Started in Pittsburgh, Pa., in 1970 as Swanson Analysis Systems, Inc. (SASI) by John Swanson (and eventually renamed), Ansys serves the CAE (Computer Aided Engineering)/multiphysics engineering simulation market. Read more…

Choosing the Right GPU for LLM Inference and Training

December 11, 2023

Accelerating the training and inference processes of deep learning models is crucial for unleashing their true potential and NVIDIA GPUs have emerged as a game- Read more…

Intel’s Server and PC Chip Development Will Blur After 2025

January 15, 2024

Intel's dealing with much more than chip rivals breathing down its neck; it is simultaneously integrating a bevy of new technologies such as chiplets, artificia Read more…

Baidu Exits Quantum, Closely Following Alibaba’s Earlier Move

January 5, 2024

Reuters reported this week that Baidu, China’s giant e-commerce and services provider, is exiting the quantum computing development arena. Reuters reported � Read more…

Leading Solution Providers

Contributors

Comparing NVIDIA A100 and NVIDIA L40S: Which GPU is Ideal for AI and Graphics-Intensive Workloads?

October 30, 2023

With long lead times for the NVIDIA H100 and A100 GPUs, many organizations are looking at the new NVIDIA L40S GPU, which it’s a new GPU optimized for AI and g Read more…

Shutterstock 1179408610

Google Addresses the Mysteries of Its Hypercomputer 

December 28, 2023

When Google launched its Hypercomputer earlier this month (December 2023), the first reaction was, "Say what?" It turns out that the Hypercomputer is Google's t Read more…

AMD MI3000A

How AMD May Get Across the CUDA Moat

October 5, 2023

When discussing GenAI, the term "GPU" almost always enters the conversation and the topic often moves toward performance and access. Interestingly, the word "GPU" is assumed to mean "Nvidia" products. (As an aside, the popular Nvidia hardware used in GenAI are not technically... Read more…

Shutterstock 1606064203

Meta’s Zuckerberg Puts Its AI Future in the Hands of 600,000 GPUs

January 25, 2024

In under two minutes, Meta's CEO, Mark Zuckerberg, laid out the company's AI plans, which included a plan to build an artificial intelligence system with the eq Read more…

Google Introduces ‘Hypercomputer’ to Its AI Infrastructure

December 11, 2023

Google ran out of monikers to describe its new AI system released on December 7. Supercomputer perhaps wasn't an apt description, so it settled on Hypercomputer Read more…

China Is All In on a RISC-V Future

January 8, 2024

The state of RISC-V in China was discussed in a recent report released by the Jamestown Foundation, a Washington, D.C.-based think tank. The report, entitled "E Read more…

Intel Won’t Have a Xeon Max Chip with New Emerald Rapids CPU

December 14, 2023

As expected, Intel officially announced its 5th generation Xeon server chips codenamed Emerald Rapids at an event in New York City, where the focus was really o Read more…

IBM Quantum Summit: Two New QPUs, Upgraded Qiskit, 10-year Roadmap and More

December 4, 2023

IBM kicks off its annual Quantum Summit today and will announce a broad range of advances including its much-anticipated 1121-qubit Condor QPU, a smaller 133-qu Read more…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire