Google is releasing an open source harassment filter for journalists

Google’s Jigsaw unit is releasing the code for an open source anti-harassment tool called Harassment Manager. The tool, intended for journalists and other public figures, employs Jigsaw’s Perspective API to let users sort through potentially abusive comments on social media platforms starting with Twitter. It’s debuting as source code for developers to build on, then being launched as a functional application for Thomson Reuters Foundation journalists in June.

Harassment Manager can currently work with Twitter’s API to combine moderation options — like hiding tweet replies and muting or blocking accounts — with a bulk filtering and reporting system. Perspective checks messages’ language for levels of “toxicity” based on elements like threats, insults, and profanity. It sorts messages into queues on a dashboard, where users can address them in batches rather than individually through Twitter’s default moderation tools. They can choose to blur the text of the messages while they’re doing it, so they don’t need to read each one, and they can search for keywords in addition to using the automatically generated queues.

A picture of the Harassment Manager dashboard as described in the post


Harassment Manager also lets users download a standalone report containing abusive messages; this creates a paper trail for their employer or, in the case of illegal content like direct threats, law enforcement. For now, however, there’s not a standalone application that users can download. Instead, developers can freely build apps that incorporate its functionality and services using it will be launched by partners like the Thomson Reuters Foundation.

Jigsaw announced Harassment Manager on International Women’s Day, and it framed the tool as particularly relevant to female journalists who face gender-based abuse, highlighting input from “journalists and activists with large Twitter presences” as well as nonprofits like the International Women’s Media Foundation and the Committee To Protect Journalists. In a Medium post, the team says it’s hoping developers can tailor it for other at-risk social media users. “Our hope is that this technology provides a resource for people who are facing harassment online, especially female journalists, activists, politicians and other public figures, who deal with disproportionately high toxicity online,” the post reads.

A screenshot of the reporting option in Jigsaw’s Harassment Manager

Google has harnessed Perspective for automated moderation before. In 2019 it released a browser extension called Tune that let social media users avoid seeing messages with a high chance of being toxic, and it’s been used by many commenting platforms (including Vox Media’s Coral) to supplement human moderation. But as we noted around the release of Perspective and Tune, the language analysis model has historically been far from perfect. It sometimes misclassifies satirical content or fails to detect abusive messages, and Jigsaw-style AI can inadvertently associate terms like “blind” or “deaf” — which aren’t necessarily negative — with toxicity. Jigsaw itself has also been criticized for a toxic workplace culture, although Google has disputed the claims.

Unlike AI-powered moderation on services like Twitter and Instagram, however, Harassment Manager isn’t a platform-side moderation feature. It’s apparently a sorting tool for helping manage the sometimes overwhelming scale of social media feedback, something that could be relevant for people far outside the realm of journalism — even if they can’t use it for now.

Repost: Original Source and Author Link


Lapsus$ hackers breached T-Mobile’s systems and stole its source code

The Lapsus$ hacking group stole T-Mobile’s source code in a series of breaches that took place in March, as first reported by Krebs on Security. T-Mobile confirmed the attack in a statement to The Verge, and says the “systems accessed contained no customer or government information or other similarly sensitive information.”

In copies of private messages obtained by Krebs, the Lapsus$ hacking group discussed targeting T-Mobile in the week prior to the arrest of seven of its teenage members. After purchasing employees’ credentials online, the members could use the company’s internal tools — like Atlas, T-Mobile’s customer management system — to perform SIM swaps. This type of attack involves hijacking a target’s mobile phone by transferring its number to a device owned by the attacker. From there, the attacker can obtain texts or calls received by that person’s phone number, including any messages sent for multi-factor authentication.

According to screenshotted messages posted by Krebs, Lapsus$ hackers also attempted to crack into the FBI and Department of Defense’s T-Mobile accounts. They were ultimately unable to do so, as additional verification measures were required.

“Several weeks ago, our monitoring tools detected a bad actor using stolen credentials to access internal systems that house operational tools software,” T-Mobile said in an emailed statement to The Verge. “Our systems and processes worked as designed, the intrusion was rapidly shut down and closed off, and the compromised credentials used were rendered obsolete.”

T-Mobile has been the victim of several attacks over the years. Although this particular hack didn’t affect customers’ data, past incidents did. In August 2021, a breach exposed the personal information belonging to over 47 million customers, while another attack occurring just months later compromised “a small number” of customer accounts.

Lapsus$ has made a name for itself as a hacking group that primarily targets the source code of large technology companies, like Microsoft, Samsung, and Nvidia. The group, which is reportedly led by a teenage mastermind, has also targeted Ubisoft, Apple Health partner Globant, and authentication company Okta.

Repost: Original Source and Author Link


Microsoft confirms Lapsus$ hackers stole source code via ‘limited’ access

The hacking group Lapsus$, known for claiming to have hacked Nvidia, Samsung, and more, this week claimed it has even hacked Microsoft. The group posted a file that it claimed contains partial source code for Bing and Cortana in an archive holding nearly 37GB of data.

On Tuesday evening, after investigating, Microsoft confirmed the group that it calls DEV-0537 compromised “a single account” and stole parts of source code for some of its products. A blog post on its security site says Microsoft investigators have been tracking the Lapsus$ group for weeks, and details some of the methods they’ve used to compromise victims’ systems. According to the Microsoft Threat Intelligence Center (MSTIC), “the objective of DEV-0537 actors is to gain elevated access through stolen credentials that enable data theft and destructive attacks against a targeted organization, often resulting in extortion. Tactics and objectives indicate this is a cybercriminal actor motivated by theft and destruction.”

Microsoft maintains that the leaked code is not severe enough to cause an elevation of risk, and that its response teams shut down the hackers mid-operation.

Lapsus$ has been on a tear recently if its claims are to be believed. The group says it’s had access to data from Okta, Samsung, and Ubisoft, as well as Nvidia and now Microsoft. While companies like Samsung and Nvidia have admitted their data was stolen, Okta pushed back against the group’s claims that it has access to its authentication service, claiming that “The Okta service has not been breached and remains fully operational.”


This week, the actor made public claims that they had gained access to Microsoft and exfiltrated portions of source code. No customer code or data was involved in the observed activities. Our investigation has found a single account had been compromised, granting limited access. Our cybersecurity response teams quickly engaged to remediate the compromised account and prevent further activity.

Microsoft does not rely on the secrecy of code as a security measure and viewing source code does not lead to elevation of risk. The tactics DEV-0537 used in this intrusion reflect the tactics and techniques discussed in this blog. Our team was already investigating the compromised account based on threat intelligence when the actor publicly disclosed their intrusion. This public disclosure escalated our action allowing our team to intervene and interrupt the actor mid-operation, limiting broader impact.

This isn’t the first time Microsoft’s claimed it assumes attackers will access its source code — it said the same thing after the Solarwinds attack. Lapsus$ also claims that it only got around 45 percent of the code for Bing and Cortana, and around 90 percent of the code for Bing Maps. The latter feels like a less valuable target than the other two, even if Microsoft was worried about its source code revealing vulnerabilities.

In its blog post, Microsoft outlines a number of steps other organizations can take to improve their security, including requiring multifactor authentication, not using “weak” multifactor authentication methods like text messages or secondary email, educating team members about the potential for social engineering attacks, and creating processes for potential responses to Lapsus$ attacks. Microsoft also says that it’ll keep tracking Lapsus$, keeping an eye on any attacks it carries out on Microsoft customers.

Repost: Original Source and Author Link


RISC-V grows open source processor membership 130% in 2021

Hear from CIOs, CTOs, and other C-level and senior execs on data and AI strategies at the Future of Work Summit this January 12, 2022. Learn more

RISC-V International said it has grown during the pandemic as its RISC-V open source processor membership popped 130% in 2021.

The nonprofit group’s membership has grown from a ragtag group of feisty academics to some of the biggest tech companies on earth like Google. Over the past decade, the group has groomed RISC-V into a viable alternative to proprietary Arm and Intel-based processors, and it appears that a lot of big companies and engineering geeks like what they see.

RISC-V chip revenues are expected to generate $400 million in 2021 and reach $1 billion in revenue by 2024, according to a prediction made this week by accounting and consulting firm Deloitte. The firm said the ripples of RISC-V could turn into the waves of the future. Calista Redmond, CEO of RISC-V International, said in an interview with VentureBeat that there are 2,4278 members in the group now, up 130% since the start of the year, and 292 companies, up 27.5% this year.

“We’ve increased and grown and we have seen a deeper investment coming in from around the world, as you see that the reflection in our board composition, which is both premier members who are coming in at our highest level of membership, as well as elected representatives of various groups,” Redmond said. “That deeper investment is a reflection of them bringing RISC-V across their portfolio of products rather than just isolated to a few projects.”

The RISC-V membership will be gathering in San Francisco both in-person and online at an event anchored at the Moscone convention center. The members are announcing today that they have ratified 15 new specifications — representing more than 40 extensions to the hardware architecture — for the RISC-V instruction set architecture, which anyone can use for free.

Calista Redmond, CEO of the RISC-V Foundation.

Above: Calista Redmond, CEO of the RISC-V Foundation.

Image Credit: RISC-V Foundation

Redmond said one of the benefits of RISC-V is that it is sanction-free. As an open source platform, RISC-V is not affected by export restrictions. This makes it appealing to companies, especially in China, that have been affected or fear being affected by those restrictions, Deloitte said. Redmond said that the nonprofit had to reboot its entire membership base as it transferred its headquarters from the U.S. to Switzerland in order to erase any doubt that it was independent of geographic borders.

Deloitte said that companies are planning on using it for different storage, graphics, and machine-learning applications. Even Intel’s foundry services division is partnering with RISC-V player SiFive. Arm argues that it has more features and has more support options for developers. Since Arm is based in the United Kingdom and Intel in the U.S., Chinese manufacturers worry that they could lose access to the architectures if trade friction heats up. Nvidia is hoping regulators will approve its plan to pay $40 billion to buy Arm.

Redmond said there are a lot of Chinese members, but overall RISC-V’s base is about a third North America, a third European, and a third Asian.

“We’ve always been global. There is nothing that changed at all in the rules, regulations, or global constructs that we participate in,” Redmond said. “Our move was primarily just to address any concerns that the landscape could change.”

Redmond said that designers don’t have to worry about constraints on what they do and that gives them freedom for innovation.

Deloitte also said that startups care about the royalty-free open source architecture. In the three years between 2020 and 2022, venture capitalists (VCs) will invest about $22 billion into startup chip companies of all kinds, Deloitte said. A million-dollar license fee may not matter to one of the world’s largest smartphone companies, but it does matter for a startup that has relatively little cash and a monthly burn rate, Deloitte said.

The served addressable market (SAM) for RISC-V in automotive alone was 4 million cores in 2020, forecast to rise to 150 million cores in 2022, and to 2.9 billion cores by 2025.

New specifications

Above: RISC-V software is expected to grow dramatically.

Image Credit: Tractica

Mark Himelstein, chief technology officer of RISC-V International, said in an interview with VentureBeat that the specifications cover vector, scalar cryptography, and hypervisor features that will keep extending the reach of RISC-V processors into new markets. Developers will find it easier to create RISC-V applications for artificial intelligence (AI) and machine learning (ML), the Internet of Things (IoT), connected and autonomous cars, data centers, and more, he said.

“The development of these specifications really showcased the incredible benefits of open collaboration across companies and geographies as members worked together to develop novel approaches for the latest computing requirements,” said Krste Asanović, chair of the RISC-V International, in a statement.

Redmond and Himelstein said that RISC-V’s advantages include that designs based on it are easy to modify. As such, they can offer greater flexibility than traditional chip designs.

“All 15 of those specifications have been ratified by the board. They all have passed acceptance criteria. And we’re very excited about that,” Himelstein said. “And we have another bunch on deck.”

The RISC-V Vector specification will help accelerate the computation of data intensive operations like ML inference for audio, vision, and voice processing. With RISC-V Vector, developers can process complex data arrays and scalar operations quickly and with low latency. The simplicity and flexibility of Vector allows companies to easily customize RISC-V solutions for a wide variety of edge computing applications from consumer IoT devices to industrial ML applications.

“The new RISC-V Vector specification will change the way people think about vector designs,” said Dave Ditzel, executive chairman of Esperanto Technologies, in a statement. “With just over 100 instructions, the extension offers a simple and elegant approach to efficiently process the latest machine learning algorithms.”

The RISC-V Hypervisor specification virtualizes supervisor-level architecture to efficiently host guest operating systems atop a type-1 or type-2 hypervisor. Virtual machine implementations require the RISC-V Hypervisor specification. The Hypervisor specification will help drive RISC-V adoption in cloud and embedded applications where virtualization is critical, such as in data centers, automotive applications, and industrial control applications. The RISC-V community has ported KVM and other open source virtual machines on top of simulators using the new specification.

The RISC-V Scalar Cryptography specification enables the acceleration of cryptographic workloads for small footprint deployments. These extensions significantly lower the barrier to entry for secure and efficient accelerated cryptography in IoT and embedded devices.

“The RISC-V Scalar Cryptography extensions allow for implementing standard cryptographic hash and block cipher algorithms that are an order of magnitude faster than using standard instructions in some cases. With RISC-V’s transparent and open approach, anyone can efficiently implement critical cryptographic algorithms in any class of CPU,” said Ben Marshall, cryptographic hardware engineer at PQShield and member of the RISC-V Technical Steering Committee, in a statement. “In addition to the performance benefits, these new extensions are very cheap to implement so companies can integrate popular cryptography algorithms in even the smallest connected devices.”


VentureBeat’s mission is to be a digital town square for technical decision-makers to gain knowledge about transformative technology and transact.

Our site delivers essential information on data technologies and strategies to guide you as you lead your organizations. We invite you to become a member of our community, to access:

  • up-to-date information on the subjects of interest to you
  • our newsletters
  • gated thought-leader content and discounted access to our prized events, such as Transform 2021: Learn More
  • networking features, and more

Become a member

Repost: Original Source and Author Link


Eclipse Foundation launches open source collaboration around software-defined vehicles

One of the world’s largest open source software (OSS) foundations, the Eclipse Foundation, this week announced an invitation to leaders in the technology sector to join in the commencement of a new working group initiative specifically focused on developing a new grade of open source software-defined vehicles.

Alongside the Eclipse Foundation are several top industry players that are joining the foundation’s open source collaborative effort including Microsoft, Red Hat, Bosch, and others.

“With digital technologies unlocking the future of accessible, sustainable, and safe transportation experiences, mobility services providers are increasingly looking to differentiate through software innovation,” said Ulrich Homann, corporate vice president and distinguished architect at Microsoft. “By standardizing the development, deployment, and management of software-defined vehicles through collaboration in the open-source space, businesses can bring tailored mobility solutions to their customers faster and can focus on innovations.”

The Eclipse Foundation’s initiative aims to provide “individuals and organizations with a mature, scalable, and business-friendly environment for open source software collaboration and innovation,” according to the foundation’s press release.

Benefits for mobility

The new working group will focus entirely on building next-generation vehicles based on open source. By open-sourcing this project, the foundation is hoping to pull solutions and innovation from the best and brightest enterprises and individuals across the globe — and doing so with an eye toward creating a strong foundation for software-defined vehicles and future mobility.

“The software-defined vehicle will play a key role in the future of mobility,” Christoph Hartung, president and chairman of embedded systems maker ETAS, said in a press release. “The explosive increase in complexity can only be mastered by working closely together as we do in this initiative.”

The foundation is focused on fostering an environment from which to pave the way for software-defined vehicles, but it doesn’t stop there. Eclipse is also looking at how both its new working group and the innovation of software-defined vehicles can be used to create robust accessibility options for people with various disabilities and physical needs.

“The transfer of personalized functionality across vehicles and brands will be eased — assume a rental car,” Sven Kappel, vice president and head of project for Bosch, told VentureBeat. “So, in the given hardware restraints, the needs of [an] impaired car user could be far faster retrieved and be met by a large developer base with lower implementation cost than classical vehicle architecture and software developing paradigms.”

A software-defined future

Software-defined vehicles have captured the attention of industry leaders, academics, and the public alike. Next-gen vehicle developers are increasingly looking to provide advanced mobility options to serve the global community, just as smart city technologies and initiatives are similarly on the rise.

The benefits from this open-sourced working group can extend beyond vehicles into other industries as well, including cloud computing and manufacturing. A similar open source-focused working initiative in another industry sector could create benefits ranging from collaborative interdisciplinary solutions to ensuring thoughtful inclusion of anticipated consumer needs early on.

As the automotive industry, like other sectors, continually pivots toward a software-defined future, interdisciplinary collaboration with open source technology will further enable innovation. Manufacturers and suppliers will be better equipped to leverage standards that make innovations available to more people — for the software-defined vehicle space, this means being able to bring customizable features to drivers and passengers at an accelerated rate, Homann explained to VentureBeat via email.

“A global open source community can leverage a wide variety of voices, which can lead to greater participation, such as contributing tools and development principles that can enhance diversity and inclusion,” Homann said.

By building and utilizing a strong, open foundation, vehicle manufacturers worldwide will be able to zero in on key differentiators for customers, like mobility services and end-user experience improvements, at the same time that they are saving both time and cost on the non-differentiating elements, such as operating systems, middleware, and communication protocols, Eclipse’s press release claims.

“Although we have extensive roots with the automotive community, a project of this scope and scale has never been attempted before,” said Mike Milinkovich, executive director of the Eclipse Foundation. “This initiative enables participants to get in at the ‘ground level’ and ensure they each have an equal voice in this project.”

The future of software-defined vehicles

The Eclipse Foundation — which has reportedly fostered more than 400 open source projects to date — is eyeing the future as it attempts an open source project unmatched to any of its previous 400. By creating an environment that it anticipates will become “an open ecosystem for deploying, configuring, and monitoring vehicle software in a secure and safe way,” and will assist with achieving a significant transformation for the industry at a large scale.

“The end goal of this project is a completely new type of automobile defined in free, open-to-anyone software that can be downloaded into an off-the-shelf chassis. Adding new features to your call will simply require a software update. An enormous first step in a new era of vehicle development,” a press release from Eclipse stated.

A transportation and logistics report released in August by the market data firm Statista projects that electronic systems will account for nearly 50% of the total price of a new car by 2030. Additionally, the report claims that even before then, by 2025 about 33% of new cars sold will be operated by an electric battery. In fact, the report predicts that within the next decade, the rise of mobility services and autonomous vehicles will launch a revolution throughout the entire auto sector.

In addition, another recent report, titled “Software-defined vehicle Research Report 2021: Architecture Trends and Industry Panorama,” points out that in order to keep up with the Joneses of the automotive industry, original equipment manufacturers (OEMs) must “open up vehicle programming to all enterprises by simplifying the development of vehicle software and increasing the frequency of updates, so as to master the ecological resources of developers.” This further underscores the Eclipse Foundation’s ultimate goal of inviting industry leaders to collaboratively build next-generation vehicles based on open source.

According to the press release, Eclipse plans to create a space fueled by “transparency, vendor-neutrality, and a shared voice” in order to ensure all participants in the open source-driven project have the opportunity to shape the future of the working group — and the very future of vehicle development itself.


VentureBeat’s mission is to be a digital town square for technical decision-makers to gain knowledge about transformative technology and transact.

Our site delivers essential information on data technologies and strategies to guide you as you lead your organizations. We invite you to become a member of our community, to access:

  • up-to-date information on the subjects of interest to you
  • our newsletters
  • gated thought-leader content and discounted access to our prized events, such as Transform 2021: Learn More
  • networking features, and more

Become a member

Repost: Original Source and Author Link


Facebook develops new method to reverse-engineer deepfakes and track their source

Deepfakes aren’t a big problem on Facebook right now, but the company continues to fund research into the technology to guard against future threats. Its latest work is a collaboration with academics from Michigan State University (MSU), with the combined team creating a method to reverse-engineer deepfakes: analyzing AI-generated imagery to reveal identifying characteristics of the machine learning model that created it.

The work is useful as it could help Facebook track down bad actors spreading deepfakes on its various social networks. This content might include misinformation but also non-consensual pornography — a depressingly common application of deepfake technology. Right now, the work is still in the research stage and isn’t ready to be deployed.

Previous studies in this area have been able to determine which known AI model generated a deepfake, but this work, led by MSU’s Vishal Asnani, goes a step further by identifying the architectural traits of unknown models. These traits, known as hyperparameters, have to be tuned in each machine learning model like parts in an engine. Collectively, they leave a unique fingerprint on the finished image that can then be used to identify its source.

Identifying the traits of unknown models is important, Facebook research lead Tal Hassner tells The Verge, because deepfake software is extremely easy to customize. This potentially allows bad actors to cover their tracks if investigators were trying to trace their activity.

Examples of deepfakes include these fake faces, generated by a well-known AI model called StyleGAN.
Image: The Verge

“Let’s assume a bad actor is generating lots of different deepfakes and uploads them on different platforms to different users,” says Hassner. “If this is a new AI model nobody’s seen before, then there’s very little that we could have said about it in the past. Now, we’re able to say, ‘Look, the picture that was uploaded here, the picture that was uploaded there, all of them came from the same model.’ And if we were able to seize the laptop or computer [used to generate the content], we will be able to say, ‘This is the culprit.’”

Hassner compares the work to forensic techniques used to identify which model of camera was used to take a picture by looking for patterns in the resulting image. “Not everybody can create their own camera, though,” he says. “Whereas anyone with a reasonable amount of experience and standard computer can cook their own model that generates deepfakes.”

Not only can the resulting algorithm fingerprint the traits of a generative model, but it can also identify which known model created an image and whether an image is a deepfake in the first place. “On standard benchmarks, we get state-of-the-art results,” says Hassner.

But it’s important to note that even these state-of-the-art results are far from reliable. When Facebook held a deepfake detection competition last year, the winning algorithm was only able to detect AI-manipulated videos 65.18 percent of the time. Researchers involved said that spotting deepfakes using algorithms is still very much an “unsolved problem.”

Part of the reason for this is that the field of generative AI is extremely active. New techniques are published every day, and it’s nearly impossible for any filter to keep up.

Those involved in the field are keenly aware of this dynamic, and when asked if publishing this new fingerprinting algorithm will lead to research that can go undetected by these methods, Hassner agrees. “I would expect so,” he says. “This is a cat and mouse game, and it continues to be a cat and mouse game.”

Repost: Original Source and Author Link


Twitch’s source code and streamer payment figures have been leaked in apparent hack

Hackers have accessed Twitch and leaked a vast amount of company data, including proprietary code, creator payouts and the “entirety of” Twitch confirmed the breach in a tweet Wednesday morning, but did not provide further details. 

On top of of the code, the attackers said they stole the the site’s mobile, desktop and console Twitch clients. It also accessed “proprietary SDKs and internal AWS services used by Twitch,” other properties like IGDB and CurseForge, an unreleased Steam competitor from Amazon Game Studios (code-named Vapour) and Twitch SOC internal red-teaming tools. It also shows creator payouts from 2019 until now, including top streamers like Nickmercs, TimTheTatMan and xQc . 

Although we haven’t verified the claim that “the entirety” of Twitch’s source code has been leaked, the files in the 126GB repository do appear to be genuine, and the payout figures for almost 2.4 million streamers seem to be present. The hackers said that the leak, which includes source code from almost 6,000 internal Github repositories, is also just “part one” of a larger release.

It doesn’t appear that information like user passwords, addresses and banking information were revealed, but that can’t be ruled out in a future drop. If you have a Twitch account, you should activate two-factor authentication so that bad actors can’t log into your account if your password has been stolen.

The group also stated that Twitch’s community is a “disgusting toxic cesspool,” so the action may be related to recent hate raids that prompted streamers to take a day off in protest. Twitch has previously said that it’s trying to stop the hate raid problem but that it wasn’t a “simple fix.” 

It’s not clear yet how attackers could have stolen such a large amount of data, especially considering that Twitch is owned by Amazon, which operates one of the largest web-hosting companies in the world.

Update (10/6/21, 11:33am ET): This post has been updated to reflect that Twitch confirmed on Wednesday that the breach took place.

All products recommended by Engadget are selected by our editorial team, independent of our parent company. Some of our stories include affiliate links. If you buy something through one of these links, we may earn an affiliate commission.

Repost: Original Source and Author Link


Neuron7 employs open source AI tools for field service across devices

The Transform Technology Summits start October 13th with Low-Code/No Code: Enabling Enterprise Agility. Register now!

Let the OSS Enterprise newsletter guide your open source journey! Sign up here. emerged from stealth this week to reveal its platform that combines various open source AI technologies to automate field service across many types of devices. The product’s promise earned the company $4.2 million in seed funding from Nexus Venture Partners and Battery Ventures.

Naturally, there’s already a fair number of organizations attempting to apply AI to a wide range of field service issues, from optimizing traffic routes to encouraging customers to engage bots rather than humans to resolve an issue.

It’s not likely AI platforms are going to replace the need for field technicians anytime soon, given all the issues that might be encountered once a device is deployed. However, AI will clearly play a significant role in enabling a limited number of field service technicians to support a much wider range of devices deployed anywhere in the world. is building a platform that consumes the recommendations created by open source AI engines and models. The aim is to make AI technologies accessible to organizations that typically don’t have the resources required to build AI models that specifically address the unique needs of a field service team, said CEO Niken Patel.

Open source AI tools

The Neuron7 platform ingests structured and unstructured data from a wide range of sources, including product and service manuals, knowledge bases, technician notes, customer relationship management (CRM) systems, and messaging systems such as Slack. It then applies various open source AI engines based on frameworks such as TensorFlow to determine how to best remediate a performance issue or an outright device failure, said Patel.

Designed as a software-as-a-service (SaaS) application, Neuron7’s goal is to make AI accessible to organizations that need to optimize field service across an increasing array of devices that require remote support by technicians, Patel said. Technicians can’t be expected to be experts on every potential issue or parameter for all those different devices — “No one can be an expert on every device,” he said.

In addition to aggregating all the data that technicians require to resolve an issue as soon as possible, Patel said, Neuron7 captures the unique knowledge and expertise of the technicians that service the devices to ultimately make the AI platform more accurate. That capability mitigates turnover issues that occur when experienced technicians leave an organization and new ones are onboarded.

Investing in service

Pricing for the Neuron7 platform is based on a subscription model, with tiers that depend on the number of data sets that need to be trained. However, Patel said the company is hoping to shift to a pricing model that is based more on the outcomes enabled by the platform.

Angel investors, early backers, and advisors of the company include Akash Palkhiwala, CFO at Qualcomm; Ashish Agarwal, CEO of Neudesic Global Services; Kintan Brahmbhatt, general manager for Amazon Podcasts; and Anand Chandrasekaran, executive vice president for Five9.

In the age of COVID-19, organizations are looking for ways to automate service management as much as possible to reduce the number of technicians they need to dispatch. Achieving that goal requires organizations to provide customer support technicians with as much relevant data as possible so they can resolve any issues remotely. The challenge is that the devices being deployed in B2C and B2B environments are becoming more complex, Patel said. As more complex devices are connected within an internet of things (IoT) application environment, the need to augment technicians with an AI platform becomes more pressing, he added.


VentureBeat’s mission is to be a digital town square for technical decision-makers to gain knowledge about transformative technology and transact.

Our site delivers essential information on data technologies and strategies to guide you as you lead your organizations. We invite you to become a member of our community, to access:

  • up-to-date information on the subjects of interest to you
  • our newsletters
  • gated thought-leader content and discounted access to our prized events, such as Transform 2021: Learn More
  • networking features, and more

Become a member

Repost: Original Source and Author Link

Tech News

Audacity open source audio editor has become spyware

One of open source software’s biggest strengths is, naturally, its openness, which brings other benefits like freedom of use, security through scrutiny, flexibility, and more. That is mostly thanks to the open source-friendly licenses these programs use, but, from time to time, someone comes along and tries to make changes that infuriate the community of users and developers. Sometimes, those changes can even be illegal. Such seems to be the fate that has befallen Audacity, one of the open source world’s most popular pieces of software that now comes under a very invasive privacy policy.

The brouhaha started just a few months ago when Audacity was bought by the Muse Group, the company behind equally popular music software like MuseScore, which is also open source, and Ultimate Guitar. So far, Audacity remains open source (and can’t really be changed into proprietary software in its current form), but that doesn’t mean that Muse Group can’t do some pretty damaging changes. Those changes come in the form of the new privacy policy that was just updated a few days ago, a policy that now allows it to collect user data.

As a desktop application with no core online functionality, Audacity never had any need to “phone home” in the first place. Now the privacy policy says that the new company does collect data and does so in a way that’s both over-arching and vague, most likely by design. For example, it says that it collects data necessary for law enforcement but doesn’t specify what kind of data is collected.

There are also questions regarding the storage of data, which is located in servers in the USA, Russia, and the European Economic Area. IP addresses, for example, are stored in an identifiable way for a day before being hashed and then stored in servers for a year. The new policy also disallows people under the age of 13 from using the software, which, as FOSS Post points out, is a violation of the GPL license that Audacity uses.

The open source community was understandably irked by these changes. Fortunately, Audacity is open source software, and it will most likely be taken by the community and forked in a different direction, perhaps with a different name. That will leave Muse Group to develop Audacity on its own instead of being able to leverage (and exploit) the open source community’s hard work.

Repost: Original Source and Author Link

Tech News

NFT of web’s source code sells for $5.4M — but contains errors (lol)

Sir Tim Berners-Lee has sold an NFT of the original source code for the world wide web for an eye-watering $5.4 million, but the winning bidder could be in for an unpleasant surprise: a security researcher has spotted errors in the code.

Up front: The NFT market exploded in 2021, bagging artists and celebrities whopping sums for digital tokens that authenticate ownership of collectibles.

Berners-Lee hopped on the bandwagon amid signs that the bubble was about to burst. He auctioned off an NFT representing this bundle of items:

  • The original archive of dated and time-stamped files containing the source code
  • A Scalable Vector Graphics (SVG) representation of the full code
  • A letter he wrote reflecting upon the code and his process of creating it
  • An animated visualization of the code being written

The winning bidder, however, may not be getting exactly what they expected: security researcher Mikko Hypponen spotted some errors in the code.

“Hold on…the www source that Sotheby is auctioning? The angle brackets are wrong!” he tweeted. “They’ve been — yes — HTML encoded from ‘< >’ to ‘&lt; &gt;’. Lol.”

Quick take: Hypponen found the errors in the animated visualization of the code.

“The NFT consists of multiple components, and the code seems to be fine everywhere else, but the video seems to have all special characters encoded,” Hypponen told TNW via email. “Such code would not work and could not be compiled. Who knows, such an error might make this thing even more collectible for collectors. ”

A developer suggested that the mistake was the result of using a web service to pretend to type the text seen in the video.

The enormous winning bid shows there’s still life left in the market. But it also further exposes what a weird market it is. Still, the NFT’s owner could find the error adds its own strange sense of value.

The NFT certainly isn’t worth $5.4 million to me — with or without buggy code. The owner will obviously prove me wrong if they resell it for a higher fee, but that’s hard for me to imagine — which might be why I’m writing about NFTs rather than investing fortunes in them. Regardless, I hope they enjoy their ludicrously expensive new certificate.

Update (2:30PM CET, July 1, 2021): Added comment from Mikko Hypponen.

Repost: Original Source and Author Link