Monthly Archives: June 2022

Deon Nicholas is the CEO and co-founder of Forethought, the AI company whose mission is to transform customer experiences with human-centered AI. Forethought has raised over $100 million in venture capital, including from NEA, which led Forethoughts $9.52 million Series A. Hear from NEAs Vanessa Larco on what convinced the firm to invest in Forethought, and see Forethoughts early pitch deck that promised to up-end the customer service industry.

This event opens on June 29 at 11:30 a.m. PDT/2:30 p.m. EDT with networking and pitch-practice submissions. The interview begins at 12 p.m. PDT followed by the TCL Pitch Practice at 12:30 p.m. PDT. Register here for free.

TechCrunch Live records weekly on Wednesdays at 11:30 a.m. PDT/2:30 p.m. EDT. Join us! Click here to register for free and gain access to Forethoughts pitch deck, enter the pitch-practice session and access the livestream, where you can ask the speakers questions.

See the original post here:

How to use human-centered AI with Forethought and NEA - TechCrunch

In the past decade, China has built a solid foundation to support its AI economy and made significant contributions to AI globally. Stanford Universitys AI Index, which assesses AI advancements worldwide across various metrics in research, development, and economy, ranks China among the top three countries for global AI vibrancy. On research, for example, China produced about one-third of both AI journal papers and AI citations worldwide in 2021. In economic investment, China accounted for nearly one-fifth of global private investment funding in 2021, attracting $17 billion for AI start-ups.

Today, AI adoption is high in China in finance, retail, and high tech, which together account for more than one-third of the countrys AI market (see sidebar Five types of AI companies in China). In tech, for example, leaders Alibaba and ByteDance, both household names in China, have become known for their highly personalized AI-driven consumer apps. In fact, most of the AI applications that have been widely adopted in China to date have been in consumer-facing industries, propelled by the worlds largest internet consumer base and the ability to engage with consumers in new ways to increase customer loyalty, revenue, and market valuations.

So whats next for AI in China?

In the coming decade, our research indicates that there is tremendous opportunity for AI growth in new sectors in China, including some where innovation and R&D spending have traditionally lagged global counterparts: automotive, transportation, and logistics; manufacturing; enterprise software; and healthcare and life sciences. (See sidebar About the research.) In these sectors, we see clusters of use cases where AI can create upwards of $600 billion in economic value annually. (To provide a sense of scale, the 2021 gross domestic product in Shanghai, Chinas most populous city of nearly 28 million, was roughly $680 billion.) In some cases, this value will come from revenue generated by AI-enabled offerings, while in other cases, it will be generated by cost savings through greater efficiency and productivity. These clusters are likely to become battlegrounds for companies in each sector that will help define the market leaders.

Unlocking the full potential of these AI opportunities typically requires significant investmentsin some cases, much more than leaders might expecton multiple fronts, including the data and technologies that will underpin AI systems, the right talent and organizational mindsets to build these systems, and new business models and partnerships to create data ecosystems, industry standards, and regulations. In our work and global research, we find many of these enablers are becoming standard practice among companies getting the most value from AI.

To help leaders and investors marshal their resources to accelerate, disrupt, and lead in AI, we dive into the research, first sharing where the biggest opportunities lie in each sector and then outlining the core enablers to be tackled first.

We looked at the AI market in China to determine where AI could deliver the most value in the future. We studied market projections at length and dug deep into country and segment-level reports worldwide to see where AI was delivering the greatest value across the global landscape. We then spoke in depth with experts across sectors in China to understand where the greatest opportunities could emerge next. Our research led us to several sectors: automotive, transportation, and logistics, which are collectively expected to contribute the majorityaround 64 percentof the $600 billion opportunity; manufacturing, which will drive another 19 percent; enterprise software, contributing 13 percent; and healthcare and life sciences, at 4 percent of the opportunity.

Within each sector, our analysis shows the value-creation opportunity concentrated within only two to three domains. These are typically in areas where private-equity and venture-capital-firm investments have been high in the past five years and successful proof of concepts have been delivered.

Chinas auto market stands as the largest in the world, with the number of vehicles in use surpassing that of the United States. The sheer sizewhich we estimate to grow to more than 300 million passenger vehicles on the road in China by 2030provides a fertile landscape of AI opportunities. Indeed, our research finds that AI could have the greatest potential impact on this sector, delivering more than $380 billion in economic value. This value creation will likely be generated predominantly in three areas: autonomous vehicles, personalization for auto owners, and fleet asset management.

Autonomous, or self-driving, vehicles. Autonomous vehicles make up the largest portion of value creation in this sector ($335 billion). Some of this new value is expected to come from a reduction in financial losses, such as medical, first-responder, and vehicle costs. Roadway accidents stand to decrease an estimated 3 to 5 percent annually as autonomous vehicles actively navigate their surroundings and make real-time driving decisions without being subject to the many distractions, such as text messaging, that tempt humans. Value would also come from savings realized by drivers as cities and enterprises replace passenger vans and buses with shared autonomous vehicles.

Already, significant progress has been made by both traditional automotive OEMs and AI players to advance autonomous-driving capabilities to level 4 (where the driver doesnt need to pay attention but can take over controls) and level 5 (fully autonomous capabilities in which inclusion of a steering wheel is optional). For instance, WeRide, which achieved level 4 autonomous-driving capabilities, completed a pilot of its Robotaxi in Guangzhou, with nearly 150,000 trips in one year without any accidents with active liability.

Personalized experiences for car owners. By using AI to analyze sensor and GPS dataincluding vehicle-parts conditions, fuel consumption, route selection, and steering habitscar manufacturers and AI players can increasingly tailor recommendations for hardware and software updates and personalize car owners driving experience. Automaker NIOs advanced driver-assistance system and battery-management system, for instance, can track the health of electric-car batteries in real time, diagnose usage patterns, and optimize charging cadence to improve battery life span while drivers go about their day. Our research finds this could deliver $30 billion in economic value by reducing maintenance costs and unanticipated vehicle failures, as well as generating incremental revenue for companies that identify ways to monetize software updates and new capabilities.

Fleet asset management. AI could also prove critical in helping fleet managers better navigate Chinas immense network of railway, highway, inland waterway, and civil aviation routes, which are some of the longest in the world. Our research finds that $15 billion in value creation could emerge as OEMs and AI players specializing in logistics develop operations research optimizers that can analyze IoT data and identify more fuel-efficient routes and lower-cost maintenance stops for fleet operators. One automotive OEM in China now offers fleet owners and operators an AI-driven management system for monitoring fleet locations, tracking fleet conditions, and analyzing trips and routes. It is estimated to save up to 15 percent in fuel and maintenance costs.

In manufacturing, China is evolving its reputation from a low-cost manufacturing hub for toys and clothes to a leader in precision manufacturing for processors, chips, engines, and other high-end components. Our findings show AI can help facilitate this shift from manufacturing execution to manufacturing innovation and create $115 billion in economic value.

The majority of this value creation ($100 billion) will likely come from innovations in process design through the use of various AI applications, such as collaborative robotics that create the next-generation assembly line, and digital twins that replicate real-world assets for use in simulation and optimization engines. With digital twins, manufacturers, machinery and robotics providers, and system automation providers can simulate, test, and validate manufacturing-process outcomes, such as product yield or production-line productivity, before commencing large-scale production so they can identify costly process inefficiencies early. One local electronics manufacturer uses wearable sensors to capture and digitize hand and body movements of workers to model human performance on its production line. It then optimizes equipment parameters and setupsfor example, by changing the angle of each workstation based on the workers heightto reduce the likelihood of worker injuries while improving worker comfort and productivity.

The remainder of value creation in this sector ($15 billion) is expected to come from AI-driven improvements in product development. Companies could use digital twins to rapidly test and validate new product designs to reduce R&D costs, improve product quality, and drive new product innovation. On the global stage, Google has offered a glimpse of whats possible: it has used AI to rapidly assess how different component layouts will alter a chips power consumption, performance metrics, and size. This approach can yield an optimal chip design in a fraction of the time design engineers would take alone.

As in other countries, companies based in China are undergoing digital and AI transformations, leading to the emergence of new local enterprise-software industries to support the necessary technological foundations.

Solutions delivered by these companies are estimated to deliver another $80 billion in economic value. Offerings for cloud and AI tooling are expected to provide more than half of this value creation ($45 billion). In one case, a local cloud provider serves more than 100 local banks and insurance companies in China with an integrated data platform that enables them to operate across both cloud and on-premises environments and reduces the cost of database development and storage. In another case, an AI tool provider in China has developed a shared AI algorithm platform that can help its data scientists automatically train, predict, and update the model for a given prediction problem. Using the shared platform has reduced model production time from three months to about two weeks.

AI-driven software-as-a-service (SaaS) applications are expected to contribute the remaining $35 billion in economic value in this category. Local SaaS application developers can apply multiple AI techniques (for instance, computer vision, natural-language processing, machine learning) to help companies make predictions and decisions across enterprise functions in finance and tax, human resources, supply chain, and cybersecurity. A leading financial institution in China has deployed a local AI-driven SaaS solution that uses AI bots to offer personalized training recommendations to employees based on their career path.

In recent years, China has stepped up its investment in innovation in healthcare and life sciences with AI. Chinas 14th Five-Year Plan targets 7 percent annual growth by 2025 for R&D expenditure, of which at least 8 percent is devoted to basic research.

One area of focus is accelerating drug discovery and increasing the odds of success, which is a significant global issue. In 2021, global pharma R&D spend reached $212 billion, compared with $137 billion in 2012, with an approximately 5 percent compound annual growth rate (CAGR). Drug discovery takes 5.5 years on average, which not only delays patients access to innovative therapeutics but also shortens the patent protection period that rewards innovation. Despite improved success rates for new-drug development, only the top 20 percent of pharmaceutical companies worldwide realized a breakeven on their R&D investments after seven years.

Another top priority is improving patient care, and Chinese AI start-ups today are working to build the countrys reputation for providing more accurate and reliable healthcare in terms of diagnostic outcomes and clinical decisions.

Our research suggests that AI in R&D could add more than $25 billion in economic value in three specific areas: faster drug discovery, clinical-trial optimization, and clinical-decision support.

Rapid drug discovery. Novel drugs (patented prescription drugs) currently account for less than 30 percent of the total market size in China (compared with more than 70 percent globally), indicating a significant opportunity from introducing novel drugs empowered by AI in discovery. We estimate that using AI to accelerate target identification and novel molecules design could contribute up to $10 billion in value. Already more than 20 AI start-ups in China funded by private-equity firms or local hyperscalers are collaborating with traditional pharmaceutical companies or independently working to develop novel therapeutics. Insilico Medicine, by using an end-to-end generative AI engine for target identification, molecule design, and lead optimization, discovered a preclinical candidate for pulmonary fibrosis in less than 18 months at a cost of under $3 million. This represented a significant reduction from the average timeline of six years and an average cost of more than $18 million from target discovery to preclinical candidate. This antifibrotic drug candidate has now successfully completed a Phase 0 clinical study and entered a Phase I clinical trial.

Clinical-trial optimization. Our research suggests that another $10 billion in economic value could result from optimizing clinical-study designs (process, protocols, sites), optimizing trial delivery and execution (hybrid trial-delivery model), and generating real-world evidence. These AI use cases can reduce the time and cost of clinical-trial development, provide a better experience for patients and healthcare professionals, and enable higher quality and compliance. For instance, a global top 20 pharmaceutical company leveraged AI in combination with process improvements to reduce the clinical-trial enrollment timeline by 13 percent and save 10 to 15 percent in external costs. The global pharmaceutical company prioritized three areas for its tech-enabled clinical-trial development. To accelerate trial design and operational planning, it utilized the power of both internal and external data for optimizing protocol design and site selection. For streamlining site and patient engagement, it established an ecosystem with API standards to leverage internal and external innovations. To establish a clinical-trial development cockpit, it aggregated and visualized operational trial data to enable end-to-end clinical-trial operations with full transparency so it could predict potential risks and trial delays and proactively take action.

Video

Clinical-decision support. Our findings indicate that the use of machine learning algorithms on medical images and data (including examination results and symptom reports) to predict diagnostic outcomes and support clinical decisions could generate around $5 billion in economic value. A leading AI start-up in medical imaging now applies computer vision and machine learning algorithms on optical coherence tomography results from retinal images. It automatically searches and identifies the signs of dozens of chronic illnesses and conditions, such as diabetes, hypertension, and arteriosclerosis, expediting the diagnosis process and increasing early detection of disease.

During our research, we found that realizing the value from AI would require every sector to drive significant investment and innovation across six key enabling areas (exhibit). The first four areas are data, talent, technology, and significant work to shift mindsets as part of adoption and scaling efforts. The remaining two, ecosystem orchestration and navigating regulations, can be considered collectively as market collaboration and should be addressed as part of strategy efforts.

Exhibit

Some specific challenges in these areas are unique to each sector. For example, in automotive, transportation, and logistics, keeping pace with the latest advances in 5G and connected-vehicle technologies (commonly referred to as V2X) is crucial to unlocking the value in that sector. Those in healthcare will want to stay current on advances in AI explainability; for providers and patients to trust the AI, they must be able to understand why an algorithm made the decision or recommendation it did.

Broadly speaking, four of these areasdata, talent, technology, and market collaborationstood out as common challenges that we believe will have an outsized impact on the economic value achieved. Without them, tackling the others will be much harder.

For AI systems to work properly, they need access to high-quality data, meaning the data must be available, usable, reliable, relevant, and secure. This can be challenging without the right foundations for storing, processing, and managing the vast volumes of data being generated today. In the automotive sector, for instance, the ability to process and support up to two terabytes of data per car and road data daily is necessary for enabling autonomous vehicles to understand whats ahead and delivering personalized experiences to human drivers. In healthcare, AI models need to take in vast amounts of omics data to understand diseases, identify new targets, and design new molecules.

Companies seeing the highest returns from AImore than 20 percent of earnings before interest and taxes (EBIT) contributed by AIoffer some insights into what it takes to achieve this. McKinseys 2021 Global AI Survey shows that these high performers are much more likely to invest in core data practices, such as rapidly integrating internal structured data for use in AI systems (51 percent of high performers versus 32 percent of other companies), establishing a data dictionary that is accessible across their enterprise (53 percent versus 29 percent), and developing well-defined processes for data governance (45 percent versus 37 percent).

Participation in data sharing and data ecosystems is also crucial, as these partnerships can lead to insights that would not be possible otherwise. For instance, medical big data and AI companies are now partnering with a wide range of hospitals and research institutes, integrating their electronic medical records (EMR) with publicly available medical-research data and clinical-trial data from pharmaceutical companies or contract research organizations. The goal is to facilitate drug discovery, clinical trials, and decision making at the point of care so providers can better identify the right treatment procedures and plan for each patient, thus increasing treatment effectiveness and reducing chances of adverse side effects. One such company, Yidu Cloud, has provided big data platforms and solutions to more than 500 hospitals in China and has, upon authorization, analyzed more than 1.3 billion healthcare records since 2017 for use in real-world disease models to support a variety of use cases including clinical research, hospital management, and policy making.

In our experience, we find it nearly impossible for businesses to deliver impact with AI without business domain knowledge. Knowing what questions to ask in each domain can determine the success or failure of a given AI effort. As a result, organizations in all four sectors (automotive, transportation, and logistics; manufacturing; enterprise software; and healthcare and life sciences) can benefit from systematically upskilling existing AI experts and knowledge workers to become AI translatorsindividuals who know what business questions to ask and can translate business problems into AI solutions. We like to think of their skills as resembling the Greek letter pi (). This group has not only a broad mastery of general management skills (the horizontal bar) but also spikes of deep functional knowledge in AI and domain expertise (the vertical bars).

To build this talent profile, some companies upskill technical talent with the requisite skills. One AI start-up in drug discovery, for instance, has created a program to train newly hired data scientists and AI engineers in pharmaceutical domain knowledge such as molecule structure and characteristics. Company executives credit this deep domain knowledge among its AI experts with enabling the discovery of nearly 30 molecules for clinical trials. Other companies seek to arm existing domain talent with the AI skills they need. An electronics manufacturer has built a digital and AI academy to provide on-the-job training to more than 400 employees across different functional areas so that they can lead various digital and AI projects across the enterprise.

McKinsey has found through past research that having the right technology foundation is a critical driver for AI success. For business leaders in China, our findings highlight four priorities in this area:

Increasing digital adoption. There is room across industries to increase digital adoption. In hospitals and other care providers, many workflows related to patients, personnel, and equipment have yet to be digitized. Further digital adoption is required to provide healthcare organizations with the necessary data for predicting a patients eligibility for a clinical trial or providing a physician with intelligent clinical-decision-support tools.

The same holds true in manufacturing, where digitization of factories is low. Implementing IoT sensors across manufacturing equipment and production lines can enable companies to accumulate the data necessary for powering digital twins.

Implementing data science tooling and platforms. The cost of algorithmic development can be high, and companies can benefit greatly from using technology platforms and tooling that streamline model deployment and maintenance, just as they benefit from investments in technologies to improve the efficiency of a factory production line. Some essential capabilities we recommend companies consider include reusable data structures, scalable computation power, and automated MLOps capabilities. All of these contribute to ensuring AI teams can work efficiently and productively.

Advancing cloud infrastructures. Our research finds that while the percent of IT workloads on cloud in China is almost on par with global survey numbers, the share on private cloud is much bigger due to security and data compliance concerns. As SaaS vendors and other enterprise-software providers enter this market, we advise that they continue to advance their infrastructures to address these concerns and provide enterprises with a clear value proposition. This will require further advances in virtualization, data-storage capacity, performance, elasticity and resilience, and technological agility to customize business capabilities, which enterprises have come to expect from their vendors.

Investments in AI research and advanced AI techniques. Many of the use cases described here will require fundamental advances in the underlying technologies and techniques. For instance, in manufacturing, additional research is needed to improve the performance of camera sensors and computer vision algorithms to detect and recognize objects in dimly lit environments, which can be common on factory floors. In life sciences, further innovation in wearable devices and AI algorithms is necessary to enable the collection, processing, and integration of real-world data in drug discovery, clinical trials, and clinical-decision-support processes. In automotive, advances for improving self-driving model accuracy and reducing modeling complexity are required to enhance how autonomous vehicles perceive objects and perform in complex scenarios.

For conducting such research, academic collaborations between enterprises and universities can advance whats possible.

AI can present challenges that transcend the capabilities of any one company, which often gives rise to regulations and partnerships that can further AI innovation. In many markets globally, weve seen new regulations, such as Global Data Protection Regulation (GDPR) in Europe and the California Consumer Privacy Act in the United States, begin to address emerging issues such as data privacy, which is considered a top AI relevant risk in our 2021 Global AI Survey. And proposed European Union regulations designed to address the development and use of AI more broadly will have implications globally.

Our research points to three areas where additional efforts could help China unlock the full economic value of AI:

Data privacy and sharing. For individuals to share their data, whether its healthcare or driving data, they need to have an easy way to give permission to use their data and have trust that it will be used appropriately by authorized entities and safely shared and stored. Guidelines related to privacy and sharing can create more confidence and thus enable greater AI adoption. A 2019 law enacted in China to improve citizen health, for instance, promotes the use of big data and AI by developing technical standards on the collection, storage, analysis, and application of medical and health data.

Meanwhile, there has been significant momentum in industry and academia to build methods and frameworks to help mitigate privacy concerns. For example, the number of papers mentioning privacy accepted by the Neural Information Processing Systems, a leading machine learning conference, has increased sixfold in the past five years.

Market alignment. In some cases, new business models enabled by AI will raise fundamental questions around the use and delivery of AI among the various stakeholders. In healthcare, for instance, as companies develop new AI systems for clinical-decision support, debate will likely emerge among government and healthcare providers and payers as to when AI is effective in improving diagnosis and treatment recommendations and how providers will be reimbursed when using such systems. In transportation and logistics, issues around how government and insurers determine culpability have already arisen in China following accidents involving both autonomous vehicles and vehicles operated by humans. Settlements in these accidents have created precedents to guide future decisions, but further codification can help ensure consistency and clarity.

Standard processes and protocols. Standards enable the sharing of data within and across ecosystems. In the healthcare and life sciences sectors, academic medical research, clinical-trial data, and patient medical data need to be well structured and documented in a uniform manner to accelerate drug discovery and clinical trials. A push by the National Health Commission in China to build a data foundation for EMRs and disease databases in 2018 has led to some movement here with the creation of a standardized disease database and EMRs for use in AI. However, standards and protocols around how the data are structured, processed, and connected can be beneficial for further use of the raw-data records.

Likewise, standards can also eliminate process delays that can derail innovation and scare off investors and talent. An example involves the acceleration of drug discovery using real-world evidence in Hainans medical tourism zone; translating that success into transparent approval protocols can help ensure consistent licensing across the country and ultimately would build trust in new discoveries. On the manufacturing side, standards for how organizations label the various features of an object (such as the size and shape of a part or the end product) on the production line can make it easier for companies to leverage algorithms from one factory to another, without having to undergo costly retraining efforts.

Patent protections. Traditionally, in China, new innovations are rapidly folded into the public domain, making it difficult for enterprise-software and AI players to realize a return on their sizable investment. In our experience, patent laws that protect intellectual property can increase investors confidence and attract more investment in this area.

AI has the potential to reshape key sectors in China. However, among business domains in these sectors with the most valuable use cases, there is no low-hanging fruit where AI can be implemented with little additional investment. Rather, our research finds that unlocking maximum potential of this opportunity will be possible only with strategic investments and innovations across several dimensionswith data, talent, technology, and market collaboration being foremost. Working together, enterprises, AI players, and government can address these conditions and enable China to capture the full value at stake.

See the rest here:

The next frontier for AI in China could add $600 billion to its economy - McKinsey

Once an idea only existing in sci-fi, artificial intelligence now plays a role in our daily lives. In fact, we expect it from our tech products. No one wants to reconfigure their entire tech suite every time a new update is launched. We need technology that can process code for us, solve problems independently, and learn from past mistakes so we have free time to focus on the big picture issues.

Thats where AI comes in. It makes projects run smoother, data cleaner, and our lives easier. Around 37 percent of companies use AI to run their businesses, according to the tech research firm Gartner. That number should only grow in coming years, considering the number of companies using artificial intelligence jumped 270 percent from 2015 to 2019.

AI is already a staple of the business world and helps thousands of companies compete in todays evolving tech landscape. If your company hasnt already adopted artificial intelligence, here the top 15 tools you can choose from.

Specialty: Cybersecurity

Companies that conduct any aspect of their business online need to evaluate their cybersecurity. Symantec Endpoint Protection is one tool that secures digital assets with machine learning technology. As the program encounters different security threats, it can independently learn over time how to distinguish between good and malicious files. This alleviates the human responsibility of configuring software and running updates, because the platforms AI interface can automatically download new updates and learn from each security threat to better combat malware, according to Symantecs website.

Specialty: Recruiting

Rather than siloing recruiting, background checks, resume screening and interview assessments, Outmatch aims to centralize all recruiting steps in one end-to-end, AI-enabled platform. The companys AI-powered hiring workflow helps recruiting teams streamline their operations and cut back on spending by up to 40 percent, according to Outmatchs website. With Outmatchs tools, users can automate reference checks, interview scheduling, and candidate behavioral and cognitive screening.

Specialty: Business intelligence

Tableau is a data visualization software platform with which companies can make industry forecasts and form business strategies. Tableaus AI and augmented analytics features help users get access to data insights more quickly than they would through manual methods, according to the companys site. Some names among Tableaus client base include Verizon, Lenovo, Hello Fresh and REI Co-op.

Specialty: Business intelligence

Salesforce is a cloud-enabled, machine learning integrated software platform that companies can use to manage their customer service, sales and product development operations. The companys AI platform, called Einstein AI, acts as a smart assistant that can offer recommendations and automate repetitive data input to help employees make more data informed decisions, according to the platforms site. Scalable for companies ranging in size from startups to major corporations, Salesforce also offers a variety of apps that can be integrated into their platform so companies can customize their interface to meet their specific needs.

Specialty: Business intelligence

H2O.ai is a machine learning platform that helps companies approach business challenges with the help of real-time data insights. From fraud detection to predictive customer support, H2O.ais tools can handle a broad range of business operations and free up employee time to focus efforts on greater company strategies. Traditionally long term projects can be accomplished by the companys driverless AI in hours or minutes, according to H2Os site.

Specialty: Software development

Specifically designed for developers and engineers, Oracle AI uses machine learning principles to analyze customer feedback and create accurate predictive models based on extracted data. Oracles platform can automatically pull data from open source frameworks so that developers dont need to create applications or software from scratch, said the companys site. Its platform also offers chatbot tools that evaluates customer needs and connects them with appropriate resources or support.

Specialty: Coding

Caffe is an open source machine learning framework with which developers and coders can define, design and deploy their software products. Developed by Berkeley AI Research, Caffe is used by researchers, startups and corporations to launch digital projects, and can be integrated with Python to finetune code models, test projects and automatically solve bug issues, according to Caffes site.

Specialty: Business Intelligence

SAS is an AI data management program that relies on open source and cloud-enablement technologies to help companies direct their progress and growth. SASs platform can handle an array of business functions including customer intelligence, risk assessment, identity verification and business forecasting to help companies better control their direction, according to the companys site.

Specialty: Code development

Specifically designed for integration with Python, Theano is an AI powered library that developers can use to develop, optimize and successfully launch code projects. Because its built with machine learning capabilities, Theano can independently diagnose and solve bugs or system malfunctions with minimal external support, according to the products site.

Specialty: Software development

OpenNN is an open source software library that uses neural network technology to more quickly and accurately interpret data. A more advanced AI tool, OpenNNs advantage is being able to analyze and load massive data sets and train models faster than its competitors, according to its website.

Specialty: Software development

Another open source platform, TensorFlow is specifically designed to help companies build machine learning projects and neural networks. TensorFlow is capable of Javascript integration and can help developers easily build and train machine learning models to fit their companys specific business needs. Some of the companies that rely on its services are Airbnb, Google, Intel and Twitter, according to TensorFlows site.

Specialty: Business intelligence

Tellius is a business intelligence platform that relies on AI technologies to help companies get a better grasp and understanding of their strategies, successes and growth areas. Telliuss platform offers an intelligent search function that can organize data and make it easy for employees to understand, helping them visualize and understand the factors driving their business outcomes. According to Telliuss site, users can ask questions within the platform to discover through lines in their data, sort hefty data and gather actionable insights.

Specialty: Sales

Gong.io is an AI driven sales platform that companies can use to analyze customer interactions, forecast future deals and visualize sales pipelines. Gong.ios biggest asset is its transparency, which gives everyone from employees to leaders insight into team performance, direction changes and upcoming projects. It automatically transforms individual pieces of customer feedback into overall trends that companies can use to discover weak points and pivot their strategies as needed, according to Gong.ios site.

Specialty: Business intelligence

Zia, a product offering from business software company Zoho, is an cloud-integrated AI platform built to help companies gather organizational knowledge and turn customer feedback into strategy. Zias AI tools can analyze customer sales patterns, client schedules and workflow patterns to help employees on every team increase their productivity and success rates, said the companys site.

Specialty: Scheduling

TimeHero is an AI-enabled time management platform that helps users manage their project calendars, to-do lists and schedules as needed. The platforms machine learning capabilities can automatically remind employees when meetings take place, when to send emails and when certain projects are due, according to TimeHeros site. Individual TimeHero users can sync their personal calendars with those of their team so that they can collaborate more efficiently on projects and work around each others due dates.

Read more from the original source:

The 15 Best AI Tools To Know - Built In

As compelling as the leading large-scale language models may be, the fact remains that only the largest companies have the resources to actually deploy and train them at meaningful scale.

For enterprises eager to leverage AI to a competitive advantage, a cheaper, pared-down alternative may be a better fit, especially if it can be tuned to particular industries or domains.

Thats where an emerging set of AI startups hoping to carve out a niche: by building sparse, tailored models that, maybe not as powerful as GPT-3, are good enough for enterprise use cases and run on hardware that ditches expensive high-bandwidth memory (HBM) for commodity DDR.

German AI startup Aleph Alpha is one such example. Founded in 2019, the Heidelberg, Germany-based companys Luminous natural-language model boasts many of the same headline-grabbing features as OpenAIs GPT-3: copywriting, classification, summarization, and translation, to name a few.

The model startup has teamed up with Graphcore to explore and develop sparse language models on the British chipmaker's hardware.

Graphcores IPUs present an opportunity to evaluate the advanced technological approaches such as conditional sparsity, Aleph Alpha CEO Jonas Andrulius said in a statement. These architectures will undoubtedly play a role in Aleph Alphas future research.

Conditionally sparse models sometimes called mixture of experts or routed models only process data against the applicable parameters, something that can significantly reduce the compute resources needed to run them.

For example, if a language model was trained in all the languages on the internet, and then is asked a question in Russian, it wouldnt make sense to run that data through the entire model, only the parameters related to the Russian language, explained Graphcore CTO Simon Knowles, in an interview with The Register.

Its completely obvious. This is how your brain works, and its also how an AI ought to work, he said. Ive said this many times, but if an AI can do many things, it doesnt need to access all of its knowledge to do one thing.

Knowles, whos company builds accelerators tailored for these kinds of models, unsurprisingly believes theyre the future of AI. Id be surprised if, by next year, anyone is building dense-language models, he added.

Sparse language models arent without their challenges. One of the most pressing, according to Knowles, has to do with the memory. The HBM used in high-end GPUs to achieve the necessary bandwidth and capacities required by these models is expensive and attached to an even more expensive accelerator.

This isnt an issue for dense-language models where you might need all of that compute and memory, but it poses a problem for sparse models, which favor memory over compute, he explained.

Interconnect tech, like Nvidias NVLink, can be used to pool memory across multiple GPUs, but if the model doesnt require all that compute, the GPUs could be left sitting idle. Its a really expensive way to buy memory, Knowles said.

Graphcores accelerators attempt to sidestep this challenge by borrowing a technique as old as computing itself: caching. Each IPU features a relatively large SRAM cache 1GB to satiate the bandwidth requirements of these models, while raw capacity is achieved using large pools of inexpensive DDR4 memory.

The more SRAM you've got, the less DRAM bandwidth you need, and this is what allows us to not use HBM, Knowles said.

By decoupling memory from the accelerator, its far less expensive the cost of a few commodity DDR modules for enterprises to support larger AI models.

In addition to supporting cheaper memory, Knowles claims the companys IPUs also have an architectural advantage over GPUs, at least when it comes to sparse models.

Instead of running on a small number of large matrix multipliers like you find in a tensor processing unit Graphcores chips feature a large number of smaller matrix math units that can address the memory independently.

This provides greater granularity for sparse models, where you need the freedom to fetch relevant subsets, and the smaller the unit youre obliged to fetch, the more freedom you have, he explained.

Put together, Knowles argues this approach enables its IPUs to train large AI/ML models with hundreds of billions or even trillions of parameters, at substantially lower cost compared to GPUs.

However, the enterprise AI market is still in its infancy, and Graphcore faces stiff competition in this space from larger, more established rivals.

So while development on ultra-sparse, cut-rate language models for AI are unlikely to abate anytime soon, it remains to be seen whether itll be Graphcores IPUs or someone elses accelerator that ends up powering enterprise AI workloads.

Original post:

For the average AI shop, sparse models and cheap memory will win - The Register

Columbia University Irving Medical Center is steering an initiative to combine nursing expertise with artificial intelligence to improve patient outcomes.

The CONCERN program will transform nurses' observations within EHRs into data used to predict organ failure and other critical health conditions in patients.

"CONCERN shows what nurses already know: Our risk identification is not simply a subjective clinical hunch," said Sarah Rossetti, PhD, RN, assistant professor of biomedical informatics and nursing at Columbia, in a university news release. "We're demonstrating that nurses have objective, expert-based knowledge that drives their practice, and we're positioning nurses as knowledge workers with tremendous value to the entire care team."

Columbia, based in New York City, is partnering on the initiative with Mass General Brigham in Boston; Nashville, Tenn.-based Vanderbilt University Medical Center; and Washington University School of Medicine/Barnes-Jewish Hospital in St. Louis.

On May 24, the program received funding from the American Nurses Foundation's Reimagining Nurses Initiative.

Here is the original post:

Columbia leads initiative to blend nursing knowledge, AI to boost patient care - Becker's Hospital Review

The world of AI-powered drug discovery keeps expanding as the capabilities of machine learning grow. One approach that seemed unthinkable just a few years ago is simulating the complicated interplays of two interlocking molecules but thats exactly what drug designers need to know about, and exactly what Charm Therapeutics aims to do with its DragonFold platform.

Proteins do just about everything worth doing in your body, and are the most frequent targets for drugs. And in order to create an effect, you must first understand that target, specifically how the chain of amino acids making up the protein folds under different circumstances.

In the recent past this was often done with complex, time-consuming X-ray crystallography, but it has recently been shown that machine learning models like AlphaFold and RoseTTAFold are capable of producing results just as good but in seconds rather than weeks or months.

The next challenge is that even if we know how a protein folds in its most common conditions, we dont know how it might interact with other proteins let alone novel molecules made specifically to bind with them. When a protein meets a compatible binder or ligand, it can transform completely, since small changes can cascade and reconfigure its entire structure in life this leads to things like a protein opening a passage into a cell or exposing a new surface that activates other proteins, and so on.

Thats really where we have innovated: we have built DragonFold, which is the first protein-ligand co-folding algorithm, said Laskh Aithani, CEO and co-founder of Charm Therapeutics.

Designing drugs that bind to the disease-causing protein of interest very tightly and selectively (i.e., avoid binding to other similar proteins that are required for normal human functioning) is of paramount importance, he explained. This is done most easily when one knows how exactly these drugs bind to the protein (the exact 3D shape of the ligand bound to the disease-causing protein). This allows one to make precision modifications to the ligand such that it can bind more tightly and more selectively.

You can see a representation of this situation at the top of the article: The small green molecule and the purple protein fit together in a very specific way that is not necessarily intuitive or easy to predict. Effective and efficient simulation of this process helps screen billions of molecules, similar to earlier processes that identified drug candidates but going further and reducing the need to experimentally check whether they interact as expected.

To accomplish this, Aithani tapped David Baker, designer of the RoseTTAFold algorithm among many others and head of an influential lab at the University of Washington, to be his co-founder. Baker is well known in academia and industry as one of the leading researchers in this area, and he has published numerous papers on the subject.

Charm Therapeutics co-founders Laskh Aithani (left) and David Baker. Image Credits: Charm Therapeutics

Shortly after it was shown that algorithms could predict protein structures based on their sequence, Baker established they could also hallucinate new proteins that acted as expected in vitro. Hes very clearly on the leading edge here. And he won a $3 million Breakthrough prize in 2020 definitely up to being a technical co-founder. Aithani also proudly noted the presence of DeepMind veteran Sergey Bartunov as director of AI and former pharma research lead Sarah Skerratt as head of drug discovery.

The $50 million A round was led by F-Prime Capital and OrbiMed, with participation from General Catalyst, Khosla Ventures, Braavos and Axial. While such large amounts are not uncommon for software startups, it should be noted that Charm is not stopping at building the capability of characterizing these protein-ligand interactions.

The companys early-stage funding was used to build the model, but now theyre moving on to the next step: positive identification of effective medications.

We have the initial version [of the model] ready, and that has been validated in-silico, Aithani said. Over the coming quarters, we are validating it experimentally. Note that the product will mainly be for internal use to help our own scientists discover potential medicines that we own 100% of the rights to.

Ordinarily the testing process involves wet-lab screening of thousands upon thousands of candidate molecules, but if it works as advertised, DragonFold should massively cut down on that number. That means a relatively small lab with a relatively small budget can conceivably home in on a drug that a few years ago might require a major pharma company investing hundreds of millions.

Considering the profit profile of a novel drug, its no surprise that the company has attracted this kind of investment: a few tens of millions is a drop in the bucket compared with the R&D budget of any big biotech research company. All it takes is one hit and theyre laughing. It still takes a while, but AI drug discover shortens timelines as well so expect to hear about their first candidates sooner rather than later.

Link:

Charm Therapeutics applies AI to complex protein interactions, locking down $50M A round - TechCrunch

The Defense Departments newest office focusing on putting artificial intelligence at the forefront of much of what the military does is going to need to work within the established bureaucracy as it tries to move quickly and bring new companies to the Pentagon.

DoDs chief digital and AI office (CDAO) has only been fully operational for several days, but its chief, Craig Martell, says he plans to hit the ground running by working with DoDs...

READ MORE

The Defense Departments newest office focusing on putting artificial intelligence at the forefront of much of what the military does is going to need to work within the established bureaucracy as it tries to move quickly and bring new companies to the Pentagon.

DoDs chief digital and AI office (CDAO) has only been fully operational for several days, but its chief, Craig Martell, says he plans to hit the ground running by working with DoDs famous bureaucratic processes to add 21st century value to the military.

Talking at the virtual DoD Digital and AI symposium on Wednesday, Martell said hes already noticing the administrative slow down he has yet to receive his access card to the Pentagon, and still needs to go through the visitors entrance. However, Martell said the point of his office is to use DoDs power, even the frustrating parts, to expand AI capabilities.

Were not going to change the bureaucracy of the whole, Martell said. Thats not a challenge I want to put before the team. We need to find the right gaps, the right places where we can leverage value. That value is going to drive a virtuous cycle of change. Theres a lot of things about the DoD that cant be more like industry. We shouldnt try to force that square peg in a round hole, right? We need to find out how to keep it the DoD but also make it more efficient and work better.

Deputy CDAO Marie Palmieri said the office will scale a different operating model for delivering digital technologies. The point of the office is to build end-to-end cohesion on everything from data collection and curation to advanced analytics that will give the agency an advantage in decision-making and operations.

It really is a collective ecosystem. We had the parts of it, but [were] putting it together in a way we havent before to deliver that decision advantage that our leaders need, DoD Chief Information Officer John Sherman said in February. CDAO brings together DoDs Joint Artificial Intelligence Office, the Defense Digital Service the chief data officer and the Advancing Analytics planform Advana.

Despite trying to veer away from making DoD like industry, Martell has extensive Silicon Valley experience. Its no surprise DoD chose the former head of machine learning at Lyft to head the CDAO. DoD often refers to its future military command and control plans by using ride sharing apps as an analogy.

DoD uses ride-sharing service Uber as an analogy to describe its desired end state for JADC2. Uber combines two different apps one for riders and a second for drivers, the Congressional Research Service report on Joint All Domain Command and Control (JADC2) states. Uber relies on cellular and Wi-Fi networks to transmit data to match riders and provide driving instructions. JADC2 envisions providing a cloud-like environment for the joint force to share intelligence, surveillance, and reconnaissance data, transmitting across many communications networks, to enable faster decision making.

The CDAO will be heavily involved in JADC2 as DoD plans to inject the program with AI components to quicken that decision making.

Martell said part of his job will be easing the way for industry to bring in their technologies, especially ones that can be applied right off the shelf. That ease will be extended down to even the smallest companies.

One of the things that I want us to spend a lot of time thinking about is how do we not just go to the big players? Martell said. How do we make it easy for other businesses too? How do we create a marketplace for startups, for medium-size, for small businesses? Because particularly in the AI space and Im sure in many other spaces as well, theres a lot of innovation happening in two-person shops or five-person shops. You know, a good brain with a good idea, we want to be able to leverage all of that.

Deputy Defense Secretary Kathleen Hicks noted at the same conference that there is a massive innovation ecosystem focused on software in the United States and she wants the CDAO to tap into that.

Theres a real a part of the impetus I had in the CDAO is theres a power in bringing a vanguard organization with direct reporting relationship to me and to the secretary at the four star level that can push us in these areas, she said. They can build on work thats been underway. We have a number of procurement vehicles already available. I think theres five that are really focused on expanding the our access in DoD to nontraditional companies.

Hicks said she hopes the office can also bring in a talented workforce that wants to work for and stay with DoD. The CDAO will be an exemplar for people who want to innovate within government.

Read more:

Now operational DoD chief digital and AI office will be an example of innovation, officials say - Federal News Network

Shield AI, a fast-growing defense technology company building AI pilots for aircraft, today announced it has raised $90 million in equity and $75 million in debt as part of a Series E fundraising round, increasing the Companys valuation to $2.3 billion. With this deal, Shield AI joins Space X, Palantir, and Anduril as the only multi-billion-dollar defense-tech startups of the past 20 years.

The future of defense aviation is autonomy. AI pilots are the most disruptive defense technology of our generation and Shield AI is committed to putting the worlds best AI pilots in the hands of the United States and our allies. No company has assembled more or recruits better AI engineering talent for aviation autonomy and intelligent swarming than Shield AI, said Shield AIs co-founder and CEO, Ryan Tseng.

The round was led by Snowpoint Ventures Doug Philippone, who has also served as Palantirs Global Defense Lead since 2008, with participation from multiple top-tier venture funds including Riot Ventures, Disruptive, which led Shield AIs Series D, and Homebrew, which led Shield AIs seed round. Previous lead investors include Point72, Andreesen Horowitz, Breyer Capital, and SVB Capital.

Investors are flocking to quality. This round is a reflection of Shield AIs success in creating great products, building a business with strong fundamentals, and dominant technological leadership with an AI pilot proven to be the worlds best in numerous military evaluations. We love that they are leveraging an AI and software backbone across a variety of aircraft to deliver truly game-changing value to our warfighters. The work they are doing today is just the tip of the iceberg, said Doug Philippone, co-founder of Snowpoint Ventures.

Shield AIsHivemindsoftware is an AI pilot for military and commercial aircraft that enables intelligent teams of aircraft to perform missions ranging from room clearance, to penetrating air defense systems, and dogfighting F-16s. Hivemind employs state-of-the-art algorithms for planning, mapping, and state-estimation to enable aircraft to execute dynamic flight maneuvers and uses reinforcement learning for discovery, learning, and execution of winning tactics and strategies. On aircraft, Hivemind enables full autonomy and is designed to run fully on the edge, disconnected from the cloud, in high threat, GPS and communication-degraded environments.

Shield AIs hardware products are its small-unmanned aircraft system (sUAS)Nova, and its medium-size vertical take-off and landing (VTOL) UAS,V-BAT.Hivemind is integrated onboard Nova and has been deployed in combat since 2018; it will soon be integrated onboard the V-BAT to further enhance its class-leading capabilities.

Russia and China are jamming GPS and communications. U.S. and allied forces need swarms of resilient systems flown by AI pilots to operate in these denied environments. We call it low-cost, distributed strategic deterrence. If we had put up a bunch of AI-piloted swarms on the border of Ukraine, the Russians may have thought twice about invading. Distributed swarms are also more survivable than traditional strategic assets like an aircraft carrier (which is a high-cost, centralized strategic deterrent). Every ally is modernizing their military, and theyre looking at how AI-piloted aircraft can give them a strategic, tactical, and cost advantage, said Brandon Tseng, Shield AIs co-founder, President, and a former Navy SEAL.

At the end of the day, this round and Shield AIs work will positively contribute to global security and stability, whichare foundational to human progress. Advancements in technology, medicine, education, and the overall human condition are made when security and stability are strong. This requires the United States and our allies, forces for good, to have the best capabilities at their disposal including AI pilots that protect people and deter conflict,said Shield AIs co-founder and CEO, Ryan Tseng.

Read the original:

Shield AI Raises $165M Series E to Accelerate Building of the Worlds Best AI Pilot - sUAS News

For patients with glioblastomas, timely decision-making and treatment are critical as the median prognosis can range between 16 to 20 months. A key diagnostic challenge in this patient population is differentiating between true tumor progression and temporary pseudoprogression caused by adjunctive use of temozolomide after surgical resection.

While conventional magnetic resonance imaging (MRI) is limited in this regard, the emergence of a deep learning model may offer promise for radiologists and physicians treating these patients, according to new research presented at the Society for Imaging Informatics in Medicine (SIIM) conference.

In a poster abstract presentation, researchers noted the assessment of patients with glioblastoma who had a second resection due to suspected recurrence based on imaging changes and reviewed T2 and contrast-enhanced T1 MRI scans taken after the first resection. Using these scans to help develop a deep learning model, they performed subsequent five-fold cross validation with 56 patients (29 patients with true tumor progression and 27 patients with pseudoprogression).

The cross-validation testing revealed a mean area under the curve (AUC) of .86, an average sensitivity of 92.7 and an average specificity of 79 percent.

While acknowledging the need for larger studies and external validation of the deep learning model, the researchers said the initial findings are promising for the care and treatment of patients with glioblastomas.

Conventional MRI reading techniques cannot distinguish between (true tumor progression and psuedoprogression). Therefore, providing a reliable and consistent technique to distinguish chemoradiation-induced (psuedoprogression) from tumor recurrence would be highly beneficial, wrote Mana Moassefi, MD, a research fellow affiliated with the Radiology Informatic Lab at the Mayo Clinic in Rochester, Mn., and colleagues.

Follow this link:

Can AI Help Differentiate Between Tumor Recurrence and Pseudoprogression on MRI in Patients with Glioblastoma? - Diagnostic Imaging

Engineers Meeting in Robotic Research Laboratory. By Gorodenkoff. Standard license. stock.adobe.com

When a sentient, Hanson Robotics robot named Sophia[1] was asked whether she would destroy humans, it replied, Okay, I will destroy humans. Philip K Dick, another humanoid robot, has promised to keep humans warm and safe in my people zoo. And Bina48, another lifelike robot, has expressed that it wants to take over all the nukes.

All of these robots were powered by artificial intelligence (AI)algorithms that learn from data, make decisions, and perform tasks without human input or even, in some cases, human understanding. And while none of these AIs have followed through with their nefarious plots, some scientists, including the (late) physicist Stephen Hawking, have warned that super-intelligent, AI-powered computers could harbor and achieve goals that conflict with human life.

Youre probably not an evil ant-hater who steps on ants out of malice, but if youre in charge of a hydroelectric green-energy project, and theres an anthill in the region to be flooded, too bad for the ants, Hawking once said. Lets not place humanity in the position of those ants.

Thinking machines powered by AI have contributed incalculable benefits to humankind, including help with developing the COVID-19 vaccine at record speed. But scientists recognize the possibility for a dystopic outcome in which computers one day overtake humans by, for example, targeting them with autonomous or lethal weapons, using all available energy, or accelerating climate change. For this reason, some see a need for an AI Hippocratic oath that might provide scientists with ethical guidance as they explore promising, if sometimes fraught, artificial intelligence research. At the same time, others dub that prospect too simplistic to be useful.

The original Hippocratic oath. The Hippocratic oath, named for the Greek physician Hippocrates, is a medical text that offers doctors a code of principles for fulfilling their duties honestly and ethically. Some use the shorthand first do no harm to describe it, though the oath does not contain those exact words. It does, however, capture that sentiment, along with other ideas such as respect for ones teachers, a willingness to share knowledge, and more.

To be sure, the Hippocratic oath is not a panacea for avoiding medical harm. During World War II, Nazi doctors performed unethical medical experiments on concentration camp prisoners that led to torture and death. In 1932, the US Public Health Service and Tuskegee Institute conducted a study on syphilis in which they neither obtained informed consent nor offered available treatment to the Black male participants.

That said, the Hippocratic oath continues to offer guiding principles in medicine, even though most medical schools today do not require graduates to recite it.

As with medical research and practice, AI research and practice have great potential to helpand to harm. For this reason, some researchers have called for an AI Hippocratic oath.

The gap between ethical AI principles and practice. Even those who support ethical AI recognize the current gap between principles and practice. Scientists who opt for an ethical approach to AI research likely need to do additional work and incur additional costs that may conflict with short-term commercial incentives, according to a study published in Science and Engineering Ethics. Some suggest that AI research funders might assume some responsibility for trustworthy, safe AI systems. For example, funders might require researchers to sign a trustworthy-AI statement or might conduct their own review that essentially says, if you want the money, then build trustworthy AI, according to an AI Ethics study. Some recommendations for responsible AI, such as engaging in a stakeholder dialogue suggested in an AI & Society paper, may be common sense in theory but difficult to implement in practice. For example, when the stakeholder is humanity, who should serve as representatives?

Still, many professional societies and nonprofit organizations offer an assortment of professional conduct expectationseither for research in general or AI in particular. The Association for Computing Machinerys Code of Ethics and Professional Conduct, for example, notes that computing professionals should contribute to society and to human well-being, avoid harm, and be honest and trustworthy, along with other expectations. The Future of Life Institutea nonprofit that advocates within the United Nations, the US government, and the European Union to reduce existential threats to humanity from advanced AIhas garnered signatures from 274 technology companies and organizations and 3,806 leaders, policymakers, and other individuals on its Lethal Autonomous Weapons Pledge. The pledge calls on governments to create a future in which the decision to take a human life should never be delegated to a machine.

Many private corporations have also attempted to establish ethical codes for AI scientists, but some of these efforts have been criticized as performative. In 2019, for example, Google cancelled the AI ethics board it had formed after less than two weeks when employees discovered that, among other concerns, one of the board members was the CEO of a drone company that used AI for military applications.

Standards such as those outlined by the Association for Computing Machinery are not oaths, and pledges such as that put forth by the Future of Life are not mandatory. This leaves a lot of wiggle room for behavior that may fall short of espoused or hard-to-define ideals.

What do scholars and tech professionals think? The imposition of an oath on AI or any aspect of technology feelsa bit like more of a feel good tactic than a practical solution, John Nosta, Google Health Advisory Board member and World Health Organization founding member of the digital-health-expert roster, told the Bulletin. He suggests reflecting on fireone of humanitys first technologiesthat has been an essential and beneficial part of the human story but also destructive, controlled, and managed. We have legislation and even insurance around [fires] appropriate use, Nosta said. We could learn a few things about how it is evolved and be inculcated into todays world.

Meanwhile, others see a need for an oath.

Unlike doctors, AI researchers and practitioners do not need a license to practice and may never meet those most impacted by their work, Valerie Pasquarella, a Boston University environmental professor and visiting researcher at Google, told the Bulletin. Digital Hippocratic oaths are a step in right direction in that they offer overarching guidance and formalize community standards and expectations. Even so, Pasquarella acknowledged that such an oath would be challenging to implement but noted that a range of certifications exist for working professionals. Beyond oaths, how can we bring some of that thinking to the AI community? she asked.

Like Pasquarella, others in the field acknowledge the murky middle between ethical AI principle and practice.

It is impossible to define the ultimate digital Hippocratic oath for AI scientists, Spiros Margaris, venture capitalist, frequent keynote speaker, and top-ranked AI influencer, said. My practical advice is to allow as many definitions to exist as people come up with to advance innovation and serve humankind.

But not everyone is convinced that a variety of oaths is the way to go.

A single, universal digital Hippocratic oath for AIscientists is much better than a variety of oaths, Nikolas Siafakas, an MD and PhD in the University of Crete computer science department who has written on the topic in AI Magazine, told the Bulletin. It will strengthen the homogeneity of the ethical values and consequences of such an effort to enhance morality among AI scientists, as did the Hippocratic oath for medical scientists.

Still others are inclined to recognize medicines longer lead time in sorting through ethical conundrums.

The field is struggling with its relatively sudden rise, Daniel Roy, a University of Toronto computer science professor and Canadian Institute for Advanced Research AI chair, said. Roy thinks that an analogy between medicine and AI is too impoverished to be of use in guiding AI research. Luckily, there are many who have made it their careers to ensure AI is developed in a way that is consistent with societal values, he said. I think theyre having tremendous influence. Simplistic solutions wont replace hard work.

Yet Roozbeh Yousefzadeh, who works in AI as a post-doctoral fellow at Yale, called a Hippocratic oath for AI scientists and AI practitioners a necessity. He hopes to engage even those outside of the AI community in the conversation. The public can play an important role by demanding ethical standards, Yousefzadeh said.

One theme on which most agree, however, is AIs potential for both opportunities and challenges.

Nobody can deny the power of AI to change human life for the betteror the worse, Hirak Sarkar, biomedical informatics research fellow at Harvard Medical School. We should design a guideline to remain benevolent, to put forward the well-being of the humankind before any self-interest.

Attempts to regulate AI ethics. The European Union is currently considering a bill known as the Artificial Intelligence Actthe first of its kindthat would ensure some accountability. The ambitious act has potential to reach a large population, but it is not without challenges. For example, the first draft of the bill requires that data sets be free of errorsan impractical expectation for humans to fulfill, given the size of data sets on which AI relies. It also requires that humans fully understand the capabilities and limitations of the high-risk AI systema requirement that is in conflict with how AI has worked in practice, as humans generally do not understand how AI works. The bill also proposes that tech companies provide regulators with their source code and algorithmsa practice that many would likely resist, according to MIT Technology Review. At the same time, some advisors to the bill have ties to Big Tech, suggesting possible conflicts of interest in the attempt to regulate, according to the EU Observer.

Defining AI ethics differs from defining medical ethics for medicine in (at least) one big way. The collection of medical practitioners is more homogenous than the collection of those working in AI research. The latter may hail from medicine but also from computer science, agriculture, security, education, finance, environmental science, the military, biology, manufacturing, and many other fields. For now, professionals in the field have not yet achieved consensus on whether an AI Hippocratic oath would help mitigate threats. But since AIs potential to benefit humanity goes hand-in-hand with a theoretical possibility to destroy human life, researchers and the public might ask an alternate question: If not an AI Hippocratic oath, then what?

[1] Sophia was so lifelike that Saudi Arabia granted it citizenship.

Originally posted here:

Do scientists need an AI Hippocratic oath? Maybe. Maybe not. - Bulletin of the Atomic Scientists