DeepMind sees revenue jump and turns first ever profit

Chess Game

DeepMind, the Alphabet-owned AI research lab, has turned a profit for the first time since the company was founded in 2010.

The London-based firm recorded a profit of £43.8 million in 2020 with revenue at £826 million, according to its annual results filing with Companies House.

In previous years, DeepMind have posted losses well into the hundreds of millions that Alphabet has heavily subsidised. For example, the company lost £476 million in 2019 and has made...

Google’s MUM will start adding more context to searches

Google is getting a little help from its MUM (Multitask Unified Model) to add more context to search results.

MUM was announced at Google I/O in May and aims to transform how the web giant handles complex queries. The model uses the T5 text-to-text framework and is said to be “1,000 times more powerful” than BERT (Bidirectional Encoder Representations from Transformers), which itself was a major breakthrough when it was introduced in 2018.

One example used by Google...

Study finds YouTube’s AI serves most of the content users regret viewing

A major crowdsourced study has found that YouTube’s AI is still recommending most of the videos that users regret viewing.

Mozilla launched a study using a crowdsourced army of volunteers that found, of the content the users regretted viewing, 71 percent were recommendations from YouTube’s AI.

https://twitter.com/mozilla/status/1412699327982669825

Over the years, concerns have been raised that YouTube’s algorithms have promoted videos that range from stupid...

MLCommons releases latest MLPerf Training benchmark results

Open engineering consortium MLCommons has released its latest MLPerf Training community benchmark results.

MLPerf Training is a full system benchmark that tests machine learning models, software, and hardware.

The results are split into two divisions: closed and open. Closed submissions are better for comparing like-for-like performance as they use the same reference model to ensure a level playing field. Open submissions, meanwhile, allow participants to submit a...

Google launches fully managed cloud ML platform Vertex AI

Google Cloud has launched Vertex AI, a fully managed cloud platform that simplifies the deployment and maintenance of machine learning models.

Vertex was announced during this year’s virtual I/O developer conference and somewhat breaks from Google’s tradition of using its keynote to focus more on updates to its mobile and web development solutions. Google announcing the platform during the keynote shows how important the company believes it to be for a wide range of...

Google’s AI reservation service Duplex is now available in 49 states

Google has expanded its controversial AI reservation service Duplex to 49 states across the US.

Duplex will have to comply with privacy regulations which vary between states and – when it expands further outside the US – their national laws too.

Google says the rollout delay in the US was due to awaiting changes in legislation or the need to add features on a per-state basis. Some states, for example, require a call-back number.

The reservation service...

Google is changing its paper review process following internal revolt

Google is making changes to how it reviews papers following an internal revolt over the company’s controversial practices.

Leading AI ethics researcher Timnit Gebru was fired from Google in December last year after sending an email to colleagues which criticised the company’s practices.

Gebru claims Google blocks the publication of papers that may cause criticism of the company’s work; including her most recent which questioned whether language models can be too...

‘Information gap’ between AI creators and policymakers needs to be resolved – report

An article posted by the World Economic Forum (WEF) has argued there is a 'huge gap in understanding' between policymakers and AI creators.

The report, authored by Adriana Bora, AI policy researcher and project manager at The Future Society, and David Alexandru Timis, outgoing curator at Brussels Hub, explores how to resolve accountability and trust-building issues with AI technology.

Bora and Timis note there is "a need for sound mechanisms that will generate a...

Google is leaking AI talent following ethicist’s controversial firing

Some high-profile AI experts have departed Google after the controversial firing of leading ethicist Timnit Gebru.

Gebru was fired from Google after criticising the company’s practices in an email following a dispute over a paper she was told not to publish which questioned whether language models can be too big and whether they can increase prejudice and inequalities. In her email, Gebru also expressed frustration at the lack of progress in hiring women at...

Google is telling its scientists to give AI a ‘positive’ spin

Google has reportedly been telling its scientists to give AI a “positive” spin in research papers.

Documents obtained by Reuters suggest that, in at least three cases, Google’s researchers were requested to refrain from being critical of AI technology.

A “sensitive topics” review was established by Google earlier this year to catch papers which cast a negative light on AI ahead of their publication.

Google asks its scientists to consult with legal,...