This Week In Digital Trust

This week we dive into the hype and fear around DeepSeek, the Chinese AI start-up taking the tech world by storm.We break down the apparent concerns driving the collective freakout over the app, which include bans by various national governments.We also explore the broader takeaways for Australia's AI capability and ecosystem and the responsible AI movement. Links:DeepSeek FAQ (Stratechery)https://stratechery.com/2025/deepseek-faq/Article about DeepSeek market impact (ABC News)https://www.abc.net.au/news/2025-02-02/deepseek-nvidia-financial-markets-frenzy-ai-race/104866302Matt Levine on the links between hedge funds and LLMs (Bloomberg - paywall)https://www.bloomberg.com/opinion/articles/2025-01-27/hedge-fund-ai-is-cheap-aiArticle about DeepSeek being banned from Australian government devices (Guardian)https://www.theguardian.com/technology/2025/feb/04/deepseek-banned-from-australian-government-devices-over-national-security-concernsArticle about DeepSeek censorship (Guardian)https://www.theguardian.com/technology/2025/jan/28/we-tried-out-deepseek-it-works-well-until-we-asked-it-about-tiananmen-square-and-taiwanArticle explaining how DeepSeek censorship works at a technical level (WIRED)https://www.wired.com/story/deepseek-censorship/Casey Newton analysis (Platformer)https://www.platformer.news/deepseek-ai-explainer-china-worries/Article reporting potential shift in Australia's focus on AI safety (InnovationAus - paywall)https://www.innovationaus.com/not-just-the-guardrails-chalmers-shifts-govt-focus-on-ai/Article on Australia's unique AI opportunity (InnovationAus - paywall)https://www.innovationaus.com/deepseek-unearths-australias-ai-opportunity/ Credits:Editing and post-production by Martin Franklin (East Coast Studio) www.eastcoaststudio.com.auMusic by Bensound.com

We're back in 2025. In this episode we break down Meta's decision to cease factchecking and permit dehumanising speech on its platforms. We explore the likely motivations and impacts of these changes, and the broader takeaways for Australia. Links: Article about Meta's factchecking changes (Reuters) https://www.reuters.com/technology/meta-ends-third-party-fact-checking-program-adopts-x-like-community-notes-model-2025-01-07/ Analysis of factchecking changes (Platformer) https://www.platformer.news/meta-trump-pivot-messenger-themes-labor-zuckerberg-wishlist/ Detail on Meta's change to dehumanising speech (Platformer) https://www.platformer.news/meta-new-trans-guidelines-hate-speech/?ref=platformer-newsletter Oped on factchecking changes (NYT) https://www.nytimes.com/2025/01/14/opinion/meta-fact-checking-policy.html Article on response to factchecking changes (Guardian) https://www.theguardian.com/australia-news/2025/jan/08/australia-meta-ending-factchecking-facebook-misinformation-sarah-hanson-young Meta explainer on its factchecking https://transparency.meta.com/en-gb/features/how-fact-checking-works/ Meta's marked up Hateful Conduct Policy (Meta) https://transparency.meta.com/en-gb/policies/community-standards/hateful-conduct/ Blog post on why collective understandings on social media can be dangerous (Programmable Mutter) https://www.programmablemutter.com/p/were-getting-the-social-media-crisis Blog post about Zuck's weird 'masculine energy' thing (Unmanned) https://sorayachemaly.substack.com/p/1-masculine-energy-or-the-vengeful Credits: Editing and post-production by Martin Franklin (East Coast Studio) eastcoaststudio.com.au Music by Bensound.com

For our final episode of 2024, we take an optimistic look at the year that was (after a bit of a throat clearing on the things that bugged us), and what we're looking forward to in 2025. Our conversation covers developments in AI, privacy reform, social media and the growing digital rights community in Australia. We'll be taking a break over the summer holidays and will be back in late January. Links: Essay on revitalising the internet ecosystem (NOEMA) https://www.noemamag.com/we-need-to-rewild-the-internet/ Article about the end of big tech by Signal president (WIRED) https://www.wired.com/story/the-beginning-of-the-end-of-big-tech/   Credits:Editing and post-production by Martin Franklin (East Coast Studio) eastcoaststudio.com.auMusic by Bensound.com

This week we discuss the Australian Government's plan to make 16 the minimum age to access social media. The plan has broad political support and resonates with many concerned parents, but there's more to it than meets the eye. We discuss the specifics of how the ban will be implemented (with key elements not yet solved and potentially problematic), and interrogate the underlying philosophical approach of banning social media for kids rather than seeking to make the platforms less harmful for everyone.   Links: PM media release on ban https://www.pm.gov.au/media/minimum-age-social-media-access-protect-australian-kids Article about parliamentary inquiry into social media (Guardian) Australian parliamentary inquiry stops short of backing social media ban for under-16s https://www.theguardian.com/technology/2024/nov/18/australian-parliamentary-inquiry-stops-short-of-backing-social-media-ban-for-under-16s Open letter from researchers arguing against the ban https://au.reset.tech/uploads/ACRT-Open-letter-re-social-media-bans.pdf Article about eSafety Commissioner's lack of public support (Crikey) https://www.crikey.com.au/2024/11/19/teen-social-media-ban-esafety-commissioner-julie-inman-grant/#:~:text=When%20Prime%20Minister%20Anthony%20Albanese,in%20rolling%20out%20the%20ban.   Credits:Editing and post-production by Martin Franklin (East Coast Studio) eastcoaststudio.com.auMusic by Bensound.com

Awareness is growing that the power of many AI tools derives from the fact that they are trained on our personal information. We discuss how regulators (like Australia's OAIC) are now stepping up their expectations that privacy obligations be adhered to by developers and deployers of AI. We also explore the case study of Australian start-up harrison.ai - whose AI tool was trained on hundreds of thousands of medical images - and which recently brought the issue of privacy and AI into the spotlight.   Links: OAIC privacy guidance for GenAI developers https://www.oaic.gov.au/privacy/privacy-guidance-for-organisations-and-government-agencies/guidance-on-privacy-and-developing-and-training-generative-ai-models OAIC privacy guidance for GenAI deployers https://www.oaic.gov.au/privacy/privacy-guidance-for-organisations-and-government-agencies/guidance-on-privacy-and-the-use-of-commercially-available-ai-products Crikey reporting on Harrison.ai and I-MED (Crikey) https://www.crikey.com.au/2024/09/19/patient-scan-data-train-artificial-intelligence-consent/ OAIC blog post about GenAI and personal information https://www.oaic.gov.au/news/blog/can-personal-information-be-used-to-develop-or-train-a-generative-ai-model Article about Harrison.ai (AFR - paywalled) https://www.afr.com/young-rich/the-brothers-making-a-fortune-revolutionising-healthcare-with-ai-20240905-p5k834 Credits:Editing and post-production by Martin Franklin (East Coast Studio) eastcoaststudio.com.auMusic by Bensound.com

This week on the podcast, our eyes are opened to a remarkable online advertising practice called "real-time bidding". RTB is where an auction involving the sensitive information of online users is initiated as they load up a web page, with advertisers bidding for the space on the page. Recent reports from Reset.Tech Australia and the Irish Council of Civil Liberties reveal the practice has led to the exposure of sensitive information about users - from purchasing habits through to real-time location and insights into their mental health - and is also fuelling scams and national security concerns. We discuss the broader implications of this practice, and the options for managing the risks it poses.   Links:Any Buyer Accepted (Reset.Tech Australia) https://au.reset.tech/news/any-buyer-accepted-unregulated-data-markets-create-personal-security-risks/Australia’s Hidden Security Crisis (Irish Council of Civil Liberties) https://www.iccl.ie/digital-data/australias-hidden-security-crisis/Scam Prevention Framework https://ministers.treasury.gov.au/ministers/stephen-jones-2022/media-releases/albanese-government-unveils-scams-prevention-frameworkArticle about US bill to ban sale of personal data to foreign adversaries (The Verge) https://www.theverge.com/2024/3/20/24106991/house-data-broker-foreign-adversaries-bill-passes   Credits:Editing and post-production by Martin Franklin (East Coast Studio) eastcoaststudio.com.auMusic by Bensound.com

This week on the podcast, we're looking at a promising development in helping Australian organisations manage AI risks. The Australian Government has released its Voluntary AI Safety Standard, as part of its safe and responsible AI agenda. The Standard features 10 guardrails as a practical blueprint for how Australian organisations can safely and responsibly use and innovate with AI. We provide our take on the Standard and assessment of the challenges in seeing it adopted across the economy. Links: elevenM blog on the Voluntary AI Safety Standard https://elevenm.com.au/blog/breaking-down-the-voluntary-ai-safety-standard/ Voluntary AI Safety Standard https://www.industry.gov.au/publications/voluntary-ai-safety-standard/introduction-standard SAAM - Safe AI made easy for small businesses https://www.saam.com.au/ Credits: Editing and post-production by Martin Franklin (East Coast Studio) eastcoaststudio.com.au Music by Bensound.com

This week on the podcast, we discuss how the conversation about AI risks seems to be shifting away from the catastrophic, existential, wiping-out-of-humanity type of scenarios. While the X-risk proponents are still out there, media coverage, regulators and the public at large seem to be homing in on more immediate and tangible AI concerns like discrimination, privacy violations, and misinformation – to name a few. We explore the reasons for this shift, which includes the fact that many people now have first-hand experience of many AI products - and their limitations. Links: Article about over-focus on existential risk (Scientific American) https://www.scientificamerican.com/article/we-need-to-focus-on-ais-real-harms-not-imaginary-existential-risks/ Article about ASIC trial (Crikey) https://www.crikey.com.au/2024/09/03/ai-worse-summarising-information-humans-government-trial/ Article about California AI safety bill SB 1047 https://www.politico.com/news/2024/09/29/gavin-veto-ai-safety-bill-00181583 Article about Australia's Voluntary AI Safety Standard (elevenM) https://elevenm.com.au/blog/breaking-down-the-voluntary-ai-safety-standard/ Credits: Editing and post-production by Martin Franklin (East Coast Studio) eastcoaststudio.com.au Music by Bensound.com

After five years of consultation on privacy reforms, the Federal Government has finally introduced legislation into the parliament. Its been billed by the Attorney General as the "first tranche" and a "first step", with promises of more reforms to come. In this episode, we share our reaction to the bill and break down what's in and what's been left for a later day. We also explore why comprehensive privacy reforms seem so hard to legislate, particularly as the Government seeks to introduce a suite of other tech reforms and obligations on topics such as social media, AI and digital identity. **Correction** - at around 6 minutes in, Jordan says the OAIC's new 'speeding ticket' infringement notices can be up to $62,000. Infringement notices can actually be up to $330,000 for companies. Links:Privacy and Other Legislation Amendment Bill 2024 https://parlinfo.aph.gov.au/parlInfo/search/display/display.w3p;query=Id%3A%22legislation%2Fbills%2Fr7249_first-reps%2F0000%22;rec=0#4d76d5260955459fb4aa15467f77172eAttorney General's media release https://www.markdreyfus.com/media/media-releases/better-protection-of-australians-privacy-mark-dreyfus-kc-mp/elevenM analysis of reforms https://elevenm.com.au/blog/its-time-to-uplift-privacy-why-organisations-need-to-act-now-following-new-law-reforms/Article about reforms and reaction (InnovationAus) https://www.innovationaus.com/timid-privacy-reforms-disappoint-industry-and-advocates/OAIC reaction https://www.oaic.gov.au/news/media-centre/oaic-welcomes-first-step-in-privacy-reformsKatharine Kemp reaction (The Conversation) https://theconversation.com/long-overdue-australian-privacy-law-reform-is-here-and-its-still-not-fit-for-the-digital-era-238214 CHOICE on consumer demand for reform https://www.choice.com.au/privacyreform Credits:Editing and post-production by Martin Franklin (East Coast Studio) eastcoaststudio.com.auMusic by Bensound.com

This week Arj is joined by elevenM’s resident digital ID ‘aficionado’ Brett Watson to discuss all things digital identity. In Episode 85 Jordan and Brett unpacked what a digital ID is, what it can do, and the various services and components that make up a digital ID system. Since then, the Australian Government has passed enabling legislation for its digital identity system and is currently consulting on legislative rules. With the legislation passed, the government is now shifting its focus on digital ID from theory to practice. In a recent speech to the National Press Club, the Minister for Government Services Bill Shorten was raising awareness about the benefits of a digital ID, announcing a proof of concept and pilot of a digital ID capability. In this episode Arj and Brett discuss digital ID’s (steady) progress over the last year, and explore the risks, challenges and misconceptions that are likely to be a feature of the digital ID conversation over the next 12 months. Links: A government explainer on the Digital ID legislation: https://www.digitalidsystem.gov.au/what-is-digital-id/digital-id-act-2024. The transcript of Minister Shorten’s speech at the National Press Club on 13 August 2024: https://ministers.dss.gov.au/speeches/15616. The Australian Strategic Policy Institute’s report - ‘Australia’s new digital ID system: finding the right way to implement it’: https://www.aspi.org.au/report/australias-new-digital-id-system-finding-right-way-implement-it. About Victor Dominello and the government’s myGov advisory group: https://www.innovationaus.com/dominello-to-lead-mygov-advisory-group/. About the discontinued Gov.UK Verify digital ID system: https://trustcloud.tech/blog/collapse-gov-uk-verify-serious-questions-national-id-systems/ Credits: Editing and post-production by Martin Franklin (East Coast Studio) eastcoaststudio.com.au Music by Bensound.com