Google scraps AI ethics committee following employee backlash.
At the close of March, Our Week in Digital reported that Google had implemented an advisory committee designed to offer guidance on ethical issues relating to Artificial Intelligence, automation and related technologies.
Just ten days after the announcement however, Google have been forced to scrap their AI council following a backlash from employees. The protests emerged over the appointment of a right-wing think tank leader to the panel.
More than 2000 Google workers signed a petition deploring the company’s selection of an anti-immigration and LGBT advocate. The board member which sparked the outrage is president of the Heritage Foundation, Kay Coles James. James and her foundation have close ties to right-wing policies and the Trump administration.
In the past, James has spoken out against trans rights and LGBT protections. She has also vocalised her favour for Trump’s border wall proposal and her anti-abortion rights viewpoint.
To date, more than 2,300 Google workers have signed the letter, along with prominent academics and advocates from outside the company. In detail, the petition stated that “Google is making clear that its version of ‘ethics’ values proximity to power over the wellbeing of trans people, other LGBTQ people and immigrants”, and that “such a position directly contravenes Google’s stated values”.
AI has been significantly criticised in the past for its discriminatory tendencies. Historically it has been cited not to have recognised certain minority groups; notably trans people or women of colour. Understandable then, that this particular appointment to a board designed to advocate fairness and equality should set the cat amongst the pigeons.
The Advanced Technology External Advisory Council as it had been named had already suffered some fatalities resulting from James’ appointment. Alessandro Acquisti, a privacy expert appointed to ATEAC, had already resigned by the time the board had been ended by Google. Meanwhile, University of Oxford professor Luciano Floridi defended his decision to retain his role on the council, while simultaneously denouncing James and her views.
Despite shutting down the council, Google state they remain committed to “the important issues that AI raises”, and will find “different ways of getting outside opinions on these topics”.
Although such commitments and indeed the closure of the council are admirable, the question surely needs to be asked what Google were doing making such a controversial and ill-advised appointment in the first place?
“Institute of Technology” locations named to boost skills training.
Earlier this week, the government announced the locations of 12 “Institutes of Technology” popping up across the UK; each designed to offer young people a vocational alternative to university. Here in the west country, colleges in Somerset and Swindon have been unveiled among the locations of these progressive institutes.
Three will be based in London, while sites in the West Midlands, Milton Keynes, Durham, Exeter and York will complete the dozen.
To date, vocational training has often been seen as the second rate option by the education system. Since the Education Act of 1944, more esteem has been granted to those students suited to take the formulaic academic route – A levels and university.
The notion of introducing this network of Technology Institutions is to create a high-quality route to gain skills and qualifications comparable to those that can be awarded by universities. Each location will have an industry specialism and will be supported by employers relevant to that sector. The Durham site, for example, will involve local colleges, Newcastle University and Nissan. It will hold a specialism in “digital advanced manufacturing”.
Alongside this, there are plans for a new technical qualification, the T-Level. This is set to be introduced next year. The sites are set to open from the autumn and will herald the start of the academic year. A budget of £170m has been set aside for the project.
The plans are not without their critics. Labour have termed the plans “small-scale” and declare they will not help the “overwhelming majority” of students in technical education. The shadow government also notes that the listed locations are not reflective of the country as a whole. There is no location based in the North West for example.
Mrs May, however, believes they would “end outdated perceptions” that were biased against vocational skills. She goes on to remark that there is a desire to build a system which is flexible and diverse; one which “harnesses the talents of our young people”. Education secretary Damian Hinds stated that the institutions will be the “pinnacle of technical training”.
A lot has been written about the skills gap within the tech sector in particular. Questions have been asked regarding the ability of the UK to be able to meet the demands for tech and digital professionals. Brexit and the reduced free movement of people across Europe may discourage European nationals with these skills from taking up roles here in the UK. This is especially relevant when our neighbours across the channel are making it easier for tech professionals to take up tech and digital roles. France recently extended and improved its visa protocol for professionals working within the tech sector.
We hope that the introduction of these technical institutes will go some way to address the gap. The plans may have been criticised for not being on a large enough scale, and perhaps there are legs to this argument. Despite this though, giving young people an opportunity to achieve industry recognised qualifications can only be a good thing. Baby steps are better than no steps at all.
New UK laws may force app and content censorship on Big Tech firms.
The UK is considering placing new pressures on tech giants to censor apps and websites which feature “harmful content” under new proposals announced on Monday.
Broadly speaking, the plans would target imagery and content featuring terrorism, hate speech, self-harm, and child abuse.
Policing the Big Tech infringements would be a powerful independent regulator. This ‘super regulator’ would have a wide scope of powers, and have the ability to pressure firms that run app stores — Apple and Google, for instance — to delist or deprioritise apps that feature harmful content in some way. In extreme cases, the regulator could make it hard for users to find and download the YouTube app if it was found to infringe the new rules around harmful content, for example. Similarly, the regulator may be given the power to force Google to delist or downrank certain sites from its search results, and Facebook to stop people linking out to infringing sites in their posts.
At this stage, there is a possibility that none of these proposals will make it into any final laws. It is unarguable that Brexit negotiations and the question mark hovering over Theresa May’s leadership will considerably delay these proposals making legislation.
A source from Business Insider remarked that the government was in an exploratory stage, and that, if implemented would involve heavy collaboration with the major tech firms. The government has recommended a three month consultation on its plans, during which the industry and other affected insiders can have their say.
At this point, it is unclear how these ideas would work in practice. Notably, though, a spokesman from the government has advised that it would be the Regulator and not the government who would have the final say over which apps and sites would be blocked. This is something which may console free speech activists who may be concerned over a slide toward policies that may be considered to be censorship. The same source went on to remark that the regulator would only enforce tech firms to block apps or sites in extreme cases. Examples of this would include when an infringing service fails to prevent terrorist use of their platform.
The tech news pages have been awash with stories over regulation in recent weeks, most notably surrounding the ethical questions that AI creates. Additionally, here in the UK, the House of Lords have recommended the implementation of a body to regulate “inadequate” responses to privacy and data breaches, along with anti-social content. Big Tech firms have been criticised over the power they hold within the space, and the ability they have to use their size and bottom-line to flout rules and bend regulation.
What do you think of these new white paper proposals? Do you think this is a responsible proposal or, indeed a move toward censorship? We’d love to hear your thoughts! Leave your comments below!
Google’s Wing drones approved to make public deliveries in Australia.
Wing has been granted regulatory approval from The Australian Aviation Authority following 3000 deliveries and a significant 18 month trial period.
Initially, around 100 homes in the suburbs of Canberra will have access to the service, however, in the coming months, they plan to expand the service to include homes further afield.
The service works with local businesses including coffee shops and pharmacies to deliver their products “in minutes”.
However, the access does not come without restrictions. Drones will not be able to fly over main roads and must operate within certain hours. They will also not be permitted to fly too close to people. Additionally, those ineligible homes will be given a safety briefing about interacting with the drones.
This is an exciting development for both Google and Australia. The Australian regulator, CASA, said the launch was “very likely” to be a world first, meaning that Google has beaten Amazon in the race to offer a commercial service to the general public. Despite some high profile trials in the UK and the US, Amazon are yet to launch commercially. For Australia, it has been suggested that drone deliveries could be worth as much as AU$40 million to local businesses.
Google have big aspirations. In December, Wing announced their intentions to launch a trial service in Finland, where they would offer a free 10-minute delivery service within the country’s capital.
The skies are starting to look a little crowded; it’s not just Amazon who are competing with Alphabet’s Wing to offer drone deliveries. Flytrex, an Israeli start-up used Iceland to trial a drone logistics service back in 2017, while Flirtey was making commercial drone deliveries for US convenience store 7- Eleven in 2016. Having said this though, it is likely to be Amazon who will be chasing the tail feathers of Wing most directly. A spokesperson for the E-comm giant remarked to the Associated Press last year that the company is still “committed to making our goal of delivering packages by drones in 30 minutes or less a reality.”
Although these pioneering projects are exciting, the open skies of Australia, Finland and Iceland lend themselves to successful navigation. We imagine that drone deliveries across city scapes will be a more complicated project to take on. As such it is likely to be some time before we see our daily latte or prescription landing on our front lawns across the majority of the UK. Having said that, the deep pockets of Google and Amazon, teamed with the vast ambition of some plucky start-ups are sure to make this a reality in the coming years.
Google launches AI Platform for Data Scientists.
Google have launched AI Platform, a new collaborative model making tool for data scientists. This will give AI creators a shared, end to end environment for teams to test, train and deploy their models. AI Platform launches in beta and is designed to be used by developers, data scientists, and data engineers.
This newly created AI Platform can handle custom and premade models similar to the kind found in AI Hub’s marketplace for useful and popular AI systems. Kubeflow Pipelines for automated workflows made its debut alongside AI Hub last year, and can be used with AI Platform.
Collaborative working seems to be the buzz within the data space, and has been an area of focus within recent months. DevOps tools such as ClusterOne announced back in January the close of a $2 million funding round to help data scientists and businesses automate and optimize infrastructure management.
AI Platform will operate in the Cloud Console, and will be able to handle both streaming and batch data. Google’s data warehouse BigQuery can be used to import data, while AI Platform users can train models using AutoML or Cloud Machine Learning Engine.
It was announced this week too, that Google have also upgraded Auto ML, the service for automating the creation of custom AI models.
Google have recognised that not everyone trying to implement AI is an expert and realise that this can be a journey which may be as overwhelming as it is exciting!
Google Cloud chief AI Scientist Andrew Moore commented;
“The AI Platform is the place where if you’re taking this terrifying journey from a germ of an idea of how you can use AI in your enterprise all the way through launch of a safe, reliable deployment, the AI Platform helps you move between each of those stages in a safe way,”
Indeed, To help people who aren’t data scientists deploy AI, Google also introduced new Video and Table categories for AutoML, its technology for automating the creation of automated systems. Additionally, premade models were also made generally available today, including Contact Center AI, which launched in beta last year. Google also release Retail to supply sales and retail businesses with predictive insights.
Mr Moore goes on to say;
“We have people with many different layers of expertise trying to build machine learning systems together, and we’ve made the decision that we have to support the continuum”.
AI has been identified as a key area of tech disruption over 2019 and beyond. What are your predictions for the space? We’d love to hear your thoughts!
Leave them in the comments below!