0:13 We’re excited to sponsor the Government Innovations Showcase in Canberra today! Michael Smedley and Larry Carland are at the SSW booth, ready to chat about all things AI, from governance and safety to implementation and migrations. If you’re here, drop by and say hello! 👋 #GIWFED #AI 0 vue 22/10/2026
2:04 If AI makes SaaS features easier to study and rebuild, will enterprises build rather than buy? Copyright protects code, but vendors may make apps harder to inspect. Where’s the line between learning from a product and copying it? 0 vue 21/10/2026
1:55 Would you let an AI read your messages to make it more useful? One personal-agent benchmark puts Meta’s Muse on top. But its bigger edge may be the context across WhatsApp, Instagram and Messenger. How much access would you give it? 0 vue 20/10/2026
1:10 AI doomsday brings Terminators to mind. But a cyberattack reportedly put supermarket fridges into defrost mode, destroying food. Eve Cogan and Josh Berman look at what this says about ageing systems. Which one would disrupt your organisation most? 0 vue 19/10/2026
2:28 Google says Gemini 4 Argon found a critical flaw in hospital software that could expose sensitive data. Trusted defenders get early access. Eve Cogan and Josh Berman raise the question: how do we help defenders without helping attackers? 0 vue 14/10/2026
1:55 An apparent CEO WhatsApp message and a reportedly cloned lawyer’s voice helped trigger €95m in bank transfers. About €36m is still missing. If a familiar voice isn’t proof of identity, what check should be required before moving the money? 0 vue 13/10/2026
0:13 We’re excited to sponsor the Government Innovations Showcase in Canberra today! Michael Smedley and Larry Carland are at the SSW booth, ready to chat about all things AI, from governance and safety to implementation and migrations. If you’re here, drop by and say hello! 👋 #GIWFED #AI 0 vue 22/10/2026
2:04 If AI makes SaaS features easier to study and rebuild, will enterprises build rather than buy? Copyright protects code, but vendors may make apps harder to inspect. Where’s the line between learning from a product and copying it? 0 vue 21/10/2026
1:55 Would you let an AI read your messages to make it more useful? One personal-agent benchmark puts Meta’s Muse on top. But its bigger edge may be the context across WhatsApp, Instagram and Messenger. How much access would you give it? 0 vue 20/10/2026
1:10 AI doomsday brings Terminators to mind. But a cyberattack reportedly put supermarket fridges into defrost mode, destroying food. Eve Cogan and Josh Berman look at what this says about ageing systems. Which one would disrupt your organisation most? 0 vue 19/10/2026
2:28 Google says Gemini 4 Argon found a critical flaw in hospital software that could expose sensitive data. Trusted defenders get early access. Eve Cogan and Josh Berman raise the question: how do we help defenders without helping attackers? 0 vue 14/10/2026
1:55 An apparent CEO WhatsApp message and a reportedly cloned lawyer’s voice helped trigger €95m in bank transfers. About €36m is still missing. If a familiar voice isn’t proof of identity, what check should be required before moving the money? 0 vue 13/10/2026
0:13 We’re excited to sponsor the Government Innovations Showcase in Canberra today! Michael Smedley and Larry Carland are at the SSW booth, ready to chat about all things AI, from governance and safety to implementation and migrations. If you’re here, drop by and say hello! 👋 #GIWFED #AI
2:04 If AI makes SaaS features easier to study and rebuild, will enterprises build rather than buy? Copyright protects code, but vendors may make apps harder to inspect. Where’s the line between learning from a product and copying it?
1:55 Would you let an AI read your messages to make it more useful? One personal-agent benchmark puts Meta’s Muse on top. But its bigger edge may be the context across WhatsApp, Instagram and Messenger. How much access would you give it?
1:10 AI doomsday brings Terminators to mind. But a cyberattack reportedly put supermarket fridges into defrost mode, destroying food. Eve Cogan and Josh Berman look at what this says about ageing systems. Which one would disrupt your organisation most?
2:28 Google says Gemini 4 Argon found a critical flaw in hospital software that could expose sensitive data. Trusted defenders get early access. Eve Cogan and Josh Berman raise the question: how do we help defenders without helping attackers?
1:55 An apparent CEO WhatsApp message and a reportedly cloned lawyer’s voice helped trigger €95m in bank transfers. About €36m is still missing. If a familiar voice isn’t proof of identity, what check should be required before moving the money?
2:58 What changes when an AI agent controls a laser? Ulysses Maclaren and Levi Jackson discuss Claude learning to align one, and what physical equipment means for AI safety. 0 vue 16/09/2026
1:49 If no one has used the model in the real world yet, what do its benchmark scores actually tell us? Meta’s latest model is posting impressive numbers for speed, price, and capability. But benchmarks are only one part of the picture. Are benchmarks enough, or do you need real-world experience first? 0 vue 16/09/2026
2:14 Are we at the tipping point for real AGI? Ulysses Maclaren explores two signs: AI helping with alignment research, and AI systems contributing to the development of newer AI systems. So where should humans stay in the loop? Want more AI news? www.youtube.com/@SSWTV/shorts #AGI #AISafety 0 vue 16/09/2026
1:31 What should an AI agent be allowed to access, and who decides when to cut it off? Do you think NVIDIA’s OpenShell, would give you enough confidence to let an agent work, or would you want tighter limits? Watch Adam and Eve's discussion: www.youtube.com/shorts/uUSh2... 0 vue 06/10/2026
1:56 What would make a personal AI assistant genuinely useful in your day-to-day work? Eve Cogan shares her first impression of Dots, a personal AI agent that connects to calendar appointments, contacts, and inbox tasks. Watch the SSW TV discussion: www.youtube.com/shorts/7BlsZ... #dots 0 vue 06/10/2026
2:06 How much should an organisation share about an escaped AI agent before it has all the answers? Wait too long, and others may miss signals that could help them respond. Share too early, and you risk spreading incomplete information. For more AI News follow www.youtube.com/@SSWTV/shorts 0 vue 07/10/2026