OpenAI quietly deletes ban on using ChatGPT for "military and warfare" | New Australian laws to curb danger of artificial intelligence | Japan successfully launches an intelligence-gathering satellite
Good morning. It's Monday 15th January.
The Daily Cyber & Tech Digest focuses on the topics we work on, including cybersecurity, critical technologies, foreign interference & disinformation.
Follow us on Twitter and on LinkedIn.
OpenAI this week quietly deleted language expressly prohibiting the use of its technology for military purposes from its usage policy, which seeks to dictate how powerful and immensely popular tools like ChatGPT can be used. The Intercept
Legislation will govern the use of artificial intelligence in high-risk settings such as law enforcement, job recruitment and healthcare, with the Albanese government poised to legislate mandatory safety requirements in its first steps towards regulating the technology. The Sydney Morning Herald
Japan successfully launched a rocket carrying a government intelligence-gathering satellite Friday on a mission to watch movements at military sites in North Korea and to improve responses to natural disasters. Associated Press
World
Google to help build the first subsea cable directly connecting South America with Asia-Pacific
TechCrunch
Paul Sawers
Google is set to build a new subsea cable connecting Chile with Australia, via French Polynesia — the first such cable to directly connect South America with Asia-Pacific. Dubbed “Humboldt,” after German polymath and explorer Alexander von Humboldt, the new cable is the latest in more than a dozen similar subsea cables that Google has invested in over the past 15 years, a journey that kicked off in 2010 with Unity that stretched some 6,000 miles across the Pacific Ocean from California to Japan.
Australia
New laws to curb danger of high-risk artificial intelligence
The Sydney Morning Herald
Lisa Visentin
Legislation will govern the use of artificial intelligence in high-risk settings such as law enforcement, job recruitment and healthcare, with the Albanese government poised to legislate mandatory safety requirements in its first steps towards regulating the technology. Industry and Science Minister Ed Husic said a new advisory body would work with government, industry and academic experts to devise the best legislative framework and define what constituted high-risk technologies and their applications across industries.
ASIC focus on enforcement role for cyber security, AI
The Australian
Angelica Snowden
The role of directors and how effectively companies manage fallout from breaches is being eyed by the corporate watchdog as it considers its task in enforcing and regulating cyber security, while it also works to understand how banks and insurers are using artificial intelligence. While cyber is not a 2024 enforcement priority, Australian Securities and Investments Commission deputy chairwoman Sarah Court told The Australian it remains a focus area – particularly through the lens of directors’ duties.
Data hack hit PM’s department, RBA, AusPost
The Australian
Ellen Whinnett
Prime Minister Anthony Albanese’s department, the Reserve Bank and Australia Post are among the government agencies that had sensitive data stolen by Russia-linked hackers who compromised the servers of law firm HWL Ebsworth.
The 'Hi Mum' scam cost Australians millions. Now, it's had an upgrade
SBS
Jessica Bahr
In 2022, the 'Hi Mum' text message became one of the most common scams in Australia, with victims losing millions to perpetrators pretending to be their children in need of help. The scam was simple but effective, with over 11,000 incidents reported in Australia that year. Now, Australians are being warned the scheme has become more sophisticated, and now features AI voice impersonation.
TikTok returns fire over Liberal senator’s ‘war’
The Australian
Geoff Chambers and Simon Benson
Chinese-owned social media giant TikTok has accused the Coalition of waging a war against its users after opposition home affairs spokesman James Paterson threatened to name and shame Australian companies who use its data tracking code. The Liberal frontbencher has written to prominent Australian companies asking whether they are still using the TikTok Pixel, when they ceased using the code and if they have sought legal advice to continue using the tracking technology. Senator Paterson has given companies until Friday to respond and warned “I intend to publish responses, or lack thereof”.
Prominent Native Title body confirms 'malicious' cyber attack
National Indigenous Times
David Prestipino
One of Western Australia's most prominent Native Title bodies covering the Pilbara and Geraldton regions has denied covering up a "malicious criminal attack" that resulted in a serious data breach in late 2022. A media statement from Yamatji Marlpa Aboriginal Corporation CEO, Simon Hawkins, in December 2022 confirmed the email accounts of employees had been compromised.
China
Communist China clamps down on chatter about Taiwan’s election
Bloomberg
China’s ruling Communist Party is reining in internet discussion about Taiwan’s election, a sign of the autocratic government’s displeasure with the democratic exercise. A hashtag about the pivotal vote was trending on Chinese social media platform Weibo on Saturday morning, as polls opened. By midday, searches for “Taiwan election” yielded a noticed reading: “According to relevant laws, regulations and policies, the content of this topic is not displayed.”
China raises concerns with US over chip-making export controls, sanctions
Reuters
Joe Cash
China's Commerce Minister Wang Wentao expressed concern over U.S. curbs preventing third countries from exporting lithography machines to China during a phone call with U.S. Commerce Secretary Gina Raimondo on Thursday, his ministry said.
$3 billion top Asia fund bets on Tencent again despite China gaming rules
Bloomberg
Youkyung Lee
A fund that outperformed most of its peers has added Tencent Holdings Ltd. back to its portfolio recently, betting on the gaming company’s attractive valuation despite further industry curbs by the Chinese government. Federated Hermes Asia Ex-Japan Equity Fund, which beat 83% of its peers for the past three years, made the purchase in the new year, even after China released a draft rule on gaming restrictions in December.
USA
Lawmakers announce creation of bipartisan group on AI
The Hill
Lauren Irwin
A bipartisan group of lawmakers announced Thursday that they would be creating a “Working Group” on artificial intelligence. The AI Working Group will explore how AI “is impacting the financial services and housing industries,” including how companies use the technology in decisionmaking, development of new products, fraud prevention, compliance and how it affects the workforce, the announcement said.
US House China panel calls Intel, Nvidia and Micron heads to testify
Financial Times
Demetri Sevastopulo
The House of Representatives’ China committee has asked the chief executives of US chipmakers to testify before Congress, as the panel intensifies its scrutiny of companies with interests in China. The committee this week sent letters to Intel, Nvidia and Micron summoning their chief executives to testify, according to several people familiar with the situation.
Americas
Canadian Cyber Centre now ranks threats with SecurityScorecard solution
IT World Canada
Howard Solomon
The Canadian government’s cyber authority has started using a U.S. company’s security ratings platform to rank cyber threats to the country’s critical infrastructure.The Canadian Centre for Cyber Security said Thursday it has contracted to use SecurityScorecard’s security ratings platform. Under an arrangement with the company, the scoring will help the Cyber Centre educate critical infrastructure owner-operators on the risks facing their organizations, assisting them in remediating and measuring cybersecurity risks.
North Asia
Japan successfully launches an intelligence-gathering satellite to watch for North Korean missiles
Associated Press
Mari Yamaguchi
Japan successfully launched a rocket carrying a government intelligence-gathering satellite Friday on a mission to watch movements at military sites in North Korea and to improve responses to natural disasters. The H2A rocket, launched by Mitsubishi Heavy Industries Ltd., lifted off from the Tanegashima Space Center in southwestern Japan, carrying the optical satellite as part of Tokyo’s reconnaissance effort to rapidly build up its military capability.
China’s meddling in Taiwan election presages year of misinformation threats
The Washington Post
Joseph Menn, Naomi Nix, Cat Zakrzewski and Pranshu Verma
A sophisticated Chinese Communist Party effort to tip Saturday’s election in Taiwan may establish a template for interfering elsewhere ahead of a wave of critical global elections, analysts in multiple countries said.
UK
UK government to publish ‘tests’ on whether to pass new AI laws
Financial Times
Cristina Criddle and Anna Gross
The UK government is set to publish a series of tests that need to be met to pass new laws on artificial intelligence, as it continues to resist creating a tougher regulatory regime for the fast-developing technology. British ministers are preparing to publish criteria in the coming weeks on the circumstances in which they would enact curbs on powerful AI models created by leading companies such as OpenAI and Google, according to multiple people familiar with the impending move.
Back UK creative sector or gamble on AI, Getty Images boss tells Sunak
The Guardian
Alex Hern
Rishi Sunak needs to decide whether he wants to back the UK’s creative industries or gamble everything on an artificial intelligence boom, the chief executive of Getty Images has said. Craig Peters, who has led the image library since 2019, spoke out amid growing anger from the creative and media sector at the harvesting of their material for “training data” for AI companies. His company is suing a number of AI image generators in the UK and US for copyright infringement.
Middle East
Israel defends itself against Hague genocide allegations with Google ads
VICE
Jules Roscoe
A Google search for the ongoing hearing against Israel at the International Court of Justice yielded an ad calling South Africa’s genocide case against it “meaningless” and linking to an Israeli government website for some users on Thursday. Motherboard viewed the ad, which appears above news coverage of the hearing being held in the Hague in the Netherlands. Google told Motherboard it reviewed the ad against its policies and did not take any action.
Turkey tightens internet censorship ahead of elections
Financial Times
Adam Samson
Turkey is tightening its censorship of the internet months before an important election, highlighting concerns that President Recep Tayyip Erdoğan’s government is further restricting civil liberties. Documents seen by the Financial Times show that Turkey’s Information Technologies and Communications Authority told internet service providers a month ago to curtail access to more than a dozen popular virtual private network services.
Big Tech
Substack’s woes deepen as tech blog leaves over Nazi content
The Washington Post
Taylor Lorenz and Will Oremus
One of the newsletter platform Substack’s most prominent writers is abandoning the platform over its decision not to moderate praise for Nazis and pro-Holocaust material. Casey Newton, a technology journalist and founder of Platformer, announced the decision in his newsletter on Thursday, after weeks of back-and-forth with the company over its bare-bones content moderation policy.
Why Platformer is leaving Substack
Platformer
Casey Newton
After much consideration, we have decided to move Platformer off of Substack. Over the next few days, the publication will migrate to a new website powered by the nonprofit, open-source publishing platform Ghost. If you already subscribe to Platformer and wish to continue receiving it, you don’t need to do anything: your account will be ported over to the new platform.
X reinstated 6,103 banned accounts in Australia including 194 barred for hateful conduct
The Guardian
Josh Butler
Elon Musk’s X social media platform let go 80% of its safety engineers and reinstated thousands of banned accounts in Australia, including nearly 200 previously barred for hateful conduct, in changes the country’s eSafety commissioner labelled a “perfect storm” for online abuse.
Elon Musk’s Tesla loses US$94 billion in market valuation as winter sets in for global electric vehicle market
South China Morning Post
Tesla had a blockbuster 2023, as its shares more than doubled in 12 months. But 2024 is starting on a different note, with Elon Musk’s electric vehicle maker off to its worst start to any year – ever. The company has lost more than US$94 billion in market valuation in just the first two weeks of this year.
Artificial Intelligence
OpenAI quietly deletes ban on using ChatGPT for "military and warfare"
The Intercept
Sam Biddle
OpenAI this week quietly deleted language expressly prohibiting the use of its technology for military purposes from its usage policy, which seeks to dictate how powerful and immensely popular tools like ChatGPT can be used.
NSA official warns of hackers using AI to perfect their English in phishing schemes
NBC News
Kevin Collier
Hackers and propagandists are turning to generative artificial intelligence chatbots like ChatGPT to make their operations seem more convincing to native English speakers, a senior official at the National Security Agency said Tuesday. Speaking at the International Conference on Cyber Security at Fordham University in New York, NSA Cybersecurity Director Rob Joyce said the spy agency has seen both cybercriminals and hackers who work for foreign intelligence agencies using such chatbots to seem more like native English speakers.
AI fears creep into finance, business and law
The Washington Post
Gerrit De Vynck and David J. Lynch
Silicon Valley figures have long warned about the dangers of artificial intelligence. Now their anxiety has migrated to other halls of power: the legal system, global gatherings of business leaders and top Wall Street regulators. In the past week, the Financial Industry Regulatory Authority, the securities industry self-regulator, labeled AI an “emerging risk” and the World Economic Forum in Davos, Switzerland, released a survey that concluded AI-fueled misinformation poses the biggest near-term threat to the global economy.
Artificial Intelligence and cloud transformation ‘dominate IT investment priorities’ for 2024, according to the new Rackspace Technology survey, in association with VMware
iTWire
New research by Rackspace Technology, a leading end-to-end, hybrid multicloud technology solutions company, in association with VMware finds that despite ongoing economic uncertainty, businesses are committed to prioritising their IT investments in 2024, particularly in transformative technologies such as artificial intelligence and cloud transformation.
Research
Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training
arXiv
Evan Hubinger, Carson Denison, Jesse Mu, Mike Lambert, Meg Tong et al.
Humans are capable of strategically deceptive behavior: behaving helpfully in most situations, but then behaving very differently in order to pursue alternative objectives when given the opportunity. If an AI system learned such a deceptive strategy, could we detect it and remove it using current state-of-the-art safety training techniques? To study this question, we construct proof-of-concept examples of deceptive behavior in large language models.
Events & Podcasts
The Australian Signals Directorate from WW2 to AI: Leadership, Threats and the Future of Intelligence
Belfer Center
Please join the Belfer Center's Defense, Emerging Technology, and Strategy Program for "The Australian Signals Directorate from WW2 to AI: Leadership, Threats and the Future of Intelligence" with Rachel Noble, Director General of the Australian Signals Directorate. The session will be held on Monday, January 22nd at 12:00PM in the Belfer Library for Harvard ID holders.
The Daily Cyber & Tech Digest is brought to you by the Cyber, Technology & Security team at ASPI.