DeepMind’s AI will show off its new StarCraft 2 skills this week

DeepMind has been continuing to train its AI in the ways of StarCraft 2 and will show off its most recent progress this week.

StarCraft 2 is a complex game with many strategies, making it the perfect testing ground for AI. Google’s DeepMind first started exploring how it can use AI to beat the world’s best StarCraft players back in 2016.

In 2017, StarCraft’s developer Blizzard made 65,000 past matches available to DeepMind researchers to begin training bots....

23 January 2019 | Gaming

Talent has begun leaking from DeepMind in recent months

If DeepMind is on your CV, you could walk into most tech companies and be offered a job on the spot with a six-figure salary. The firm is full of in-demand talent and its CEO once bragged that no employees had ever left.

Speaking to The Guardian in 2016, DeepMind CEO Demis Hassabis said:

“We are able to literally get the best scientists from each country each year. So we’ll have, say, the person that won the Physics Olympiad in Poland, the person who got the top...

8 January 2019 | Deep Learning

Apple promotes AI engineer poached from Google to SVP

Sometimes it’s difficult keeping track of Silicon Valley appointments, but John Giannandrea’s journey is one worth keeping tabs on. Giannandrea is one of Silicon Valley’s brightest minds in AI. He came under the spotlight for heading Google’s search and artificial intelligence operations. After spending eight years at Google, back in April he was poached by Apple to lead its Siri and CoreML efforts. His positive impact at Apple was clearly noted. In a statement, Apple CEO Tim...

21 December 2018 | Apple

AI beats AI: AlphaZero becomes the best game player in history

DeepMind – a UK-based subsidiary of Google’s parent organisation Alphabet – has beat rival AI records with AlphaZero. The AI, introduced by DeepMind in a research published in Science on 6th November 2018, was put against three of the world’s most complex board games and their current AI record holders. This included world-champion chess AI Stockfish, DeepMind’s very own AlphaGo Zero – known as the best Go player in history – and elmo, winner of the 27th annual World Computer...

10 December 2018 | Gaming

Google funding ‘good’ AI may help some forget that military fiasco

Google has launched an initiative to fund ‘good’ AI which may help some forget about the questionable military contracts it was involved with. The new initiative, called AI for Social Good, is a joint effort between the company’s philanthropic subsidiary Google.org and its own experts. Kicking off the initiative is the ‘AI Impact Challenge’ which is set to provide $25 million in funding to non-profits while providing access to Google’s vast resources. As part of the initiative,...

30 October 2018 | Google

Experts warn of AI disasters leading to research lockdown

Experts from around the world have warned of potential AI disasters that could lead to a subsequent lockdown of research. Andrew Moore, the new head of AI at Google Cloud, is one such expert who has warned of scenarios that would lead to public backlash and restrictions that would prevent AI from reaching its full potential. Back in November, Moore spoke at the Artificial Intelligence and Global Security Initiative. In his keynote, he said:

"If an AI disaster...

13 September 2018 | Government

Harvard and Google partner on earthquake aftershock-predicting AI

Harvard University has partnered with Google to create a potentially life-saving AI that can predict where aftershocks will hit following an earthquake. The AI uses data from 199 earthquakes and the more than 130,000 aftershocks. Manual predictions are possible but they consume time and have proven to be less accurate than those of the AI. Many believe aftershocks occur within a short timeframe of an earthquake. The AI, however, can predict aftershocks up to a year after the initial...

30 August 2018 | Google

Google improves AI model training by open-sourcing framework

Google is helping researchers seeking to train AI models by open-sourcing a reinforcement learning framework used for its own projects. Reinforcement learning has been used for some of the most impressive AI demonstrations thus far, including those which beat human professional gamers at Alpha Go and Dota 2. Google subsidiary DeepMind uses it for its Deep Q-Network (DQN). Building a reinforcement learning framework takes both time and significant resources. For AI to reach its full potential,...

28 August 2018 | Google

Google is reportedly building an AI-powered fitness coach for WearOS

Google appears to be preparing an AI-powered fitness coach for WearOS as it continues its mission to revitalise the platform. Fitness remains a key reason why people buy smartwatches. Beyond that, they’re still really just an expensive place to see our smartphone notifications. WearOS’ main competitor, the Apple Watch, has produced some clever features with regards to fitness. Among the most impressive is the ability to tap some gym equipment to bring up the relevant info and tracking...

16 August 2018 | Healthcare

Google promises its call center AI is not designed to replace humans

Not content with its impressive(ly creepy) Duplex demo, Google promises it’s not wanting to replace call centers with its latest AI demonstration. During the Google Cloud Next 18 conference, Google Chief AI Scientist Dr. Fei-Fei Li demonstrated a new AI system called Google Contact Center AI which – much like Duplex – sounded incredibly natural in its responses to human queries. Google seems to have learned its lesson from its Duplex demonstration and wanted to iterate that it’s not...

25 July 2018 | Google