Cynthia Lummis, who won an open U.S. Senate seat in Wyoming, wants to explain bitcoin’s value to her new colleagues.
Read more Bitcoin is at price levels not seen since 2017, and it is outperforming ether in November.
Read more Chris Krebs, one of the most senior cybersecurity officials in the U.S. government, has been fired.
Krebs served as the director of the Cybersecurity and Infrastructure Security Agency (CISA) since its founding in November 2018 until he was removed from his position on Tuesday. It’s not immediately clear who is currently heading the agency. A spokesperson for CISA did not immediately comment.
President Trump fired Krebs in a tweet late on Tuesday, citing a statement published by CISA last week, which found there was “no evidence that any voting system deleted or lost votes, changed votes, or was in any way compromised.” Trump, who has repeatedly made claims of voter fraud without providing evidence, alleged that CISA’s statement was “highly inaccurate.”
Shortly after, Twitter labeled Trump’s tweet for making a “disputed” claim about election fraud.
Reuters first reported the news of Krebs’ potential firing last week.
Krebs was appointed by President Trump to head the newly created cybersecurity agency in November 2018, just days after the conclusion of the midterm elections. He previously served as an undersecretary for CISA’s predecessor, the National Protection and Programs Directorate, and also held cybersecurity policy roles at Microsoft.
During his time in government, Krebs became one of the most vocal voices in election security, taking the lead during 2018 and in 2020, which largely escaped from disruptive cyberattacks, thanks to efforts to prepare for cyberattacks and misinformation that plagued the 2016 presidential election.
He was “one of the few people in this administration respected by everyone on both sides of the aisle,” said Sen. Mark Warner, a member of the Senate Intelligence Committee, in a tweet.
Krebs is the latest official to leave CISA in the past year. Brian Harrell, who oversaw infrastructure protection at the agency, resigned in August after less than a year on the job, and Jeanette Manfra left for a role at Google at the end of last year. Cyberscoop reported Thursday that Bryan Ware, CISA’s assistant director for cybersecurity, resigned for a position in the private sector.
Seldon is a U.K. startup that specializes in the rarified world of development tools to optimize machine learning. What does this mean? Well, dear reader, it means that the “AI” that companies are so fond of trumpeting does actually end up working.
It has now raised a £7.1 million Series A round co-led by AlbionVC and Cambridge Innovation Capital . The round also includes significant participation from existing investors Amadeus Capital Partners and Global Brain, with follow-on investment from other existing shareholders. The £7.1 million funding will be used to accelerate R&D and drive commercial expansion, take Seldon Deploy — a new enterprise solution — to market and double the size of the team over the next 18 months.
More accurately, Seldon is a cloud-agnostic machine learning (ML) deployment specialist which works in partnership with industry leaders such as Google, Red Hat, IBM and Amazon Web Services.
Key to its success is that its open-source project Seldon Core has more than 700,000 models deployed to date, drastically reducing friction for users deploying ML models. The startup says its customers are getting productivity gains of as much as 92% as a result of utilizing Seldon’s product portfolio.
Alex Housley, CEO and founder of Seldon speaking to TechCrunch explained that companies are using machine learning across thousands of use cases today, “but the model actually only generates real value when it’s actually running inside a real-world application.”
“So what we’ve seen emerge over these last few years are companies that specialize in specific parts of the machine learning pipeline, such as training version control features. And in our case we’re focusing on deployment. So what this means is that organizations can now build a fully bespoke AI platform that suits their needs, so they can gain a competitive advantage,” he said.
In addition, he said Seldon’s open-source model means that companies are not locked-in: “They want to avoid locking as well they want to use tools from various different vendors. So this kind of intersection between machine learning, DevOps and cloud-native tooling is really accelerating a lot of innovation across enterprise and also within startups and growth-stage companies.”
Nadine Torbey, an investor at AlbionVC, added: “Seldon is at the forefront of the next wave of tech innovation, and the leadership team are true visionaries. Seldon has been able to build an impressive open-source community and add immediate productivity value to some of the world’s leading companies.”
Vin Lingathoti, partner at Cambridge Innovation Capital, said: “Machine learning has rapidly shifted from a nice-to-have to a must-have for enterprises across all industries. Seldon’s open-source platform operationalizes ML model development and accelerates the time-to-market by eliminating the pain points involved in developing, deploying and monitoring machine learning models at scale.”
The musician’s life is rarely an easy one. That goes double for these last, oh, eight or so months. Most artists who haven’t risen the ranks to rock-star levels make the lion’s share of their money from live shows, and that’s just not in the cards for now — and likely won’t be for some time.
Bandcamp has been something of a lifesaver for many artists who make fractions of a cent on streaming revenue. Among other things, the site has set up the wildly popular Bandcamp Fridays, wherein it waves its fees for one week a month. Now it’s getting into live streaming, as musicians look for ways to offer remote performances for fans.
The service is pretty well positioned to offer the feature. For one thing, it’s got goodwill going for it. For another, integration with existing services means fans get notified when a show is coming up. And all their goods are offered up for sale during the shows as a kind of virtual merch table where they can buy swag while still watching the performance.
The setup process is pretty bare bones and there’s even an optional chat on board, if the artist wants to engage in a bit of banter. Honestly, the most appealing thing here is probably the low overhead. Musicians set their own prices, there are none of the usual surprise fees and Bandcamp only takes 10% off the top — a fee it’s waiving altogether through March of next year.
The service has already lined up a bunch of high-profile indie acts, including Clap Your Hands Say Yeah, Pedro the Lion and Cloud Nothings. The feature is rolling out to users starting today.
Sight Tech Global is little more than two weeks out, and today we published the detailed agenda for Dec. 2 & 3. The show runs from 8 a.m. to noonish Pacific standard time. If you have not already grabbed a free pass to the 100% virtual event, now is the time!
Sight Tech Global will present 35 speakers in 15 sessions focused on the cutting edge of AI-related technologies and accessibility, especially for the blind and visually impaired. A few of the remarkably accomplished speakers include OrCam founder Amnon Shashua, Seeing AI co-founder Saqib Shaikh, human rights lawyer Haben Girma, computer vision researcher Danna Gurari, Amazon L126 researcher Josh Miele and AI expert and investor Kai-Fu Lee.
The agenda also includes 10 breakouts that run in parallel to the main-stage sessions. These 30-minute segments are produced by partners who are excited about the strong profile of Sight Tech Global’s 1,200+ registered attendees to date. The list of breakouts is below.
As ever, we are grateful to the excellent sponsors of Sight Tech Global, including Waymo, Salesforce, Mojo Vision, Ford, Vispero, Google, Microsoft, Amazon, Wells Fargo, Comcast, accessiBe, Eyedaptic, APH, HumanWare, Verizon Media, Verizon 5G and TechCrunch. Sponsorships benefit the nonprofit Vista Center for the Blind and Visually Impaired, which has been serving the Silicon Valley area for 75 years.
Sponsorship opportunities are still available.
Please have a look at complete agenda. Here are the breakout sessions so far!
Perkins Access: Users aren’t an add-on – Building the user perspective into the design process
Comcast and Perkins Access (the digital accessibility consulting division of Perkins School for the Blind) will share insights for creating accessible experiences, with an emphasis on building the user perspective into the design process. This ensures that all teams understand the specific challenges, and unique needs, of blind and visually impaired users. Panelists include the authors of Perkins Access’ Inclusive Design Guide, which will be released at Sight Tech and available for download.
- Gary Aussant, Director of Digital Accessibility, Perkins Access
- Geoff Freed, Director of Consulting, Perkins Access
- Jerry Berrier, Director of Education Technology, Perkins School for the Blind
- Karyn Georgilis, MBA candidate, Harvard Business School
- Tom Wlodkowski, Vice president Accessibility and Multicultural, Technology and Product, Comcast
American Council of the Blind: Get Up & get moving – A call for leveraging technology to improve health and wellness
The COVID-19 pandemic has demonstrated the global challenges that technology can experience when pushed to the limits. This realty check has not only been disproportionately burdensome for individuals who are blind and visually impaired, but it has also exposed the pre-existing barriers that have harmed the physical, social and psychological well-being within this community over the years. Join the American Council of the Blind for an empowering panel on how technology can break down barriers to a full and enriched life, and how we can all get up and get moving toward full equality in the health and wellness arena.
- Clark Rachfal, Director of Advocacy, American Council of the Blind
- Eric Bridges, Executive Director, American Council of the Blind
- Brian Charlson, member, American Council of the Blind
Benetech: Using artificial intelligence to unlock STE(A)M education
Artificial intelligence is a term that has been around for decades and AI applications and techniques are already being used in everything from HR and healthcare to e-commerce. But what is the future of AI in supporting accessibility and inclusive education? This session will provide a basic understanding of various AI techniques, including Machine Learning and Computer Vision, and how Benetech is applying these techniques to transform complex books. For accessible formats, text is easy but equations, images and other non-text content is not straightforward. Join us to hear more about the future of Assistive Technology and how it is opening new worlds for the blind and visually impaired.
- Brad Turner, VP and GM, Global Education and Literacy, Benetech
Salesforce: The new Office of Accessibility – Explained
It’s been a year since Salesforce announced the launch of their Office of Accessibility, a new corporate team that partners with internal stakeholders to highlight accessibility needs and develop improvement plans, build workforce development programs and evangelize Salesforce and their employees, customers and other important work across the industry, all under one roof.
In this breakout session, Kristian Burch, Senior Manager of Global Accessibility Compliance, and Richard Boardman, Senior Director of UX Engineering, Accessibility, will discuss what led to this groundbreaking move, how the Office interacts with other teams and more specifically Product Accessibility, what’s worked, and what they would change looking back.
- Kristian Burch, Senior Manager of Global Accessibility Compliance
- Richard Boardman, Senior Director of UX Engineering, Accessibility
Fable: The barriers to Utopia – Why feedback comes first
A lot of conversations these days are about the latest technology, and how it promises to solve all of our problems. But what about people? Join the CEO and the Community Lead of Fable, Alwar Pillai and Samuel Proulx, as they discuss how to collect authentic feedback from people living with disabilities.
- Alwar Pillai, CEO, Fable
- Samuel Proulx, Community lead
Eyedaptic: Simulated natural vision technology & one user’s low-vision journey
Eyedaptic is an AR (augmented reality) visual aid company, which helps those with retina-related vision loss, such as AMD, simulate natural vision. Eyedaptic’s novel software adapts to the user’s vision, as well as their environment and habits, and optimizes the user’s remaining vision. Samuel Newman will discuss his own low-vision challenges that he has overcome and the low-vision technologies he has tried.
- Jay Cormier, Founder and CEO, Eyedaptic
- Samuel Newman, Clinical specialist & Low-vision Technology User
Vispero: The engineering experience of adding a voice assistant to ZoomText and JAWS
Roxana and Sriram talk about their experiences in adding voice assistant to a mainstream Windows screen reader and magnifier. They explore the new input mechanic’s benefits and limitations and the guideposts they used to create the initial command set. They also talk about the voice assistant’s data and conversational privacy aspects and how Vispero is approaching them.
- Sriram Ramanathan, Senior Software Engineer, Vispero
- Roxana Fischer, Software Developer, Vispero
Humanware: Plotting the course – delving into the past, present and future of assistive technology for the visually impaired community through the lens of artificial intelligence
This session will spotlight the trajectory of HumanWare and how current technological trends impact the future of product development. Join Eric Beauchamp, François Boutrouille and Peter Tucic for a discussion of how the previous 32 years of HumanWare’s development of blindness and low-vision technology has evolved and will continue to do so with the advent of artificial intelligence and machine learning. Participants will develop a better understanding of how the challenge of providing products that solved singular tasks has now shifted to integrate the complexities of deep learning technology to interact with dynamic objectives in real time.
- Peter Tucic, Brand Ambassador of Blindness Products, HumanWare
- Eric Beauchamp, Director of Product Management, HumanWare
- François Boutrouille, Emerging Technologies Leader, HumanWare
Teach Access: Teaching accessibility to tomorrow’s builders
Teach Access, a national coalition of institutions of higher ed, corporations (mostly tech-centered) and advocates with disabilities, will be conducting a roundtable with recent college students to discuss how the teaching of accessible design and development at the university level can help close the accessibility skills gap for the emerging generation of participants in the new digital economy.
- Kate Sonka, Executive Director, Teach Access; Assistant Director of Academic Technology, Michigan State University
- Larry Goldberg, Co-founder, Teach Access; Head of Accessibility, Verizon Media
Google has taken the wraps off Chimera Painter, a web-based tool that lets anyone generate terrifying cryptozoological entities in an interface that looks like MS Paint by way of Diablo. Why, you ask? Well, isn’t it obvious? No… no, I suppose it isn’t.
Surely the strangest thing to hit Google’s AI blog for at least a month, the Chimera Painter does actually have something like a reason for existing. The team was looking at ways to accelerate the creation of art for games, which is often fantastical and creative. An AI assistant that could produce a reasonable image of, say, an owlbear on the hunt, might be helpful to an artist looking for inspiration. In 2019 Nvidia released a similar tool to generate photorealistic landscape images.
To pursue this somewhat esoteric goal, the team naturally decided to build an entire fantasy digital card game where players combine animals and make them fight. So far, I think you’ll agree this is pretty standard stuff.
Image Credits: Google
The idea was that if there are a hundred animals in the game, and each can be combined with each of the others, that quickly makes far more combinations than any artist can be expected to draw. But machine learning systems never complain, or invoice you.
To make an AI agent that can create arbitrary creatures, the team first trained it on extant animals and their many parts by feeding the system thousands of images of CG creatures and corresponding images labeling their parts: claws, front of leg, eyes, etc.
Soon the agent was able to generate plausible-looking animals from user-generated assemblies of parts, painting in fur, skin and other features according to how it had learned “real” creatures looked. It’s a generative adversarial network or GAN, which means it’s two working in concert: one generates an image, the other criticizes it, then the first takes the feedback and generates again, and so on.
Image Credits: Google
Crucially, the system doesn’t bat an eye (or should I say, dino-bat-hybrid an eye) when the assembly of labeled parts looks nothing like a real animal. For all the chimera generator knows, there are dogs with chameleon heads, long noses and tiny, useless wings. Why not?
And now, I must rescind my recent assertion that Google lacks generosity, for they have made the Chimera Painter available for all and sundry to play with. I must warn you, however, that it barely worked for me, allowing only the largest brushes, and seemingly choosing from a selection of deli meats for its different textures.
Not that it was any hindrance to the execution of my vision:
Image Credits: Devin Coldewey / Google
Splendid!
In conclusion, asks Google: “What can one create when using machine learning as a paintbrush?” Indeed, it seems there are no limits whatsoever. But perhaps there should be.
Even though the winner of an American election usually gets announced soon after the vote happens, the result is never actually official on Election Day. It’s not even official once the media makes their result projections, as happened last week. Instead, election results actually become real when state and local election authorities make sure that every valid vote was counted and formally certify them.
While President Donald Trump continues to dispute the election—he’s launched over a dozen legal efforts to prove fraud, challenge counts, and delay certification—every lawsuit has effectively failed so far.
Election officials appointed by the Trump administration say the president’s claims are dangerous and lack any credibility. Every state says there is no evidence of fraud, and federal and local election officials of both parties released a joint statement to say exactly that. As it has been since he started making accusations years ago, the president’s claims about election fraud are utterly empty.
What does certifying results really mean? The process is the same this year as it’s been any previous year. Election officials canvass results by tabulating and verifying the outcome across their states. They look at provisional ballots, and those which were challenged according to state and sometimes even county laws. After checking them over, the results are certified: the formal process in which the outcome is made official. The exact method varies state to state, but generally a secretary of state or a state board of elections will meet after counting is concluded and sign a certification of the results.
Counting may have taken longer this year because the pandemic dramatically increased the number of mail-in ballots, but the only meaningful difference is that the sitting president is carrying on an unprecedented attack on the results. Trump’s legal challenges and recount requests could theoretically alter the certification timeline, but courts have been throwing out the campaign’s efforts so far because of a lack of evidence.
That means the next couple of weeks will feature a cascade of certifications, and each will move the overall process forward. Although partisan state and local officials could theoretically block certification or appoint their own electors, there is slim to no sign that anyone currently plans to do that. In Georgia, that’s resulted in Trump lashing out at elected Republican officials unwilling to echo his lies. In Ohio, the Republican governor who co-chaired Trump’s reelection campaign acknowledged former vice president Joe Biden’s win and Trump immediately attacked him on Twitter.
Here’s what happens next in the states that have played a key role in the presidential election:
- Georgia’s certification deadline is November 20.
- Pennsylvania counties must submit certification by November 23.
- Michigan’s certification deadline is November 23.
- Nevada’s certification deadline is November 24.
- Arizona’s certification deadline is November 30.
- Wisconsin’s certification deadline is December 1.
Biden is winning in all of those states.
On December 14, the Electoral College casts its votes and then the states are done. The election then finally moves into federal hands: two months after Election Day, Congress formally elects the next president on January 6 when it counts electoral votes in a joint session.
Trying to understand Google Tag Manager? Looking for an easier way to install code on your website? In this article, you’ll discover the three parts of Google Tag Manager (tags, triggers, and variables) and learn how to use Google Tag Manager templates to easily add tracking for activities on your website and social media. You’ll […]
The post Getting Started With Google Tag Manager: A Beginner’s Guide appeared first on Social Media Examiner | Social Media Marketing.
Despite spending years preparing for the 2020 US election, insights show that Facebook remains a key facilitator of misinformation in the wake of the poll.