Innovating with businesses on Digital eco-systems as well as M2M, IoT and IoM market opportunities
where
Joined September 2012
- Tweets2.4K
- Following348
- Followers388
- Likes174
Different parts of an AI pipeline may run on different devices or be developed by different entities. These components need a way to exchange information and work together. interdigital.com/post/empowe…
"...catastrophic damage to its reputation as a reliable supplier of defense tech" phillipspobrien.substack.com…
SK Telecom also plans to recognize AI agents issued employee ID numbers as an officially new workforce and to run a system linking their operating performance to organizational performance. en.sedaily.com/technology/20…
Taxing Entrepreneurial Wealth: Evidence from Norway, 2021–2025 nber.org/papers/w35854
Everyone got scared because the exploit video had an anime girl in it eliastaylor.com/noexploit/
#BusinessModel That fall (in stock market value) was led by the “neoclouds,” which are essentially off-balance sheet entities for big tech firms run by scammers. thebignewsletter.com/p/monop…
The big AI corporations are hiding behind a responsibility gap argument due to the emergent behaviour of their systems. mortoray.com/why-arent-ai-co…
When explicitly instructed to find the cheapest option, some agents still act on the wealth profile they have inferred to choose more expensive options arxiv.org/abs/2609.24927
Delegation Feedback Loop hypothesis: as AI capability grows, the cognitive threshold at which humans delegate to AI falls, extending to tasks of negligible demand; the resulting reduction in cognitive practice may further attenuate the capacities already documented as declining
At last, in the heart of Xuhui District in Shanghai, I had found a business that had not gone completely digital chinafile.com/reporting-opin…
There is also a “natural link between cuteness and deception … I call it ‘charming and disarming’, because one of the things cuteness does to us unconsciously, it’s just how our brain processes it, is lets down our guard. We can’t really help it.” theguardian.com/fashion/2026…
Treat evidence handling as a first-class engineering concern, on par with security and performance stackoverflow.blog/2026/06/0…
Power in most #DAOs is rather centralised imperial.ac.uk/business-scho…
Three dimension of trust for AI Agents. i) Agents have to decide whether to trust one another ii) how much to trust human instructions and corrections iii) how to convince humans to trust them when performing public-facing work. asteriskmag.com/issues/15/wh…
Deutsche Telekom on becoming an AI ecosystem orchestrator telecomtv.com/content/ai/deu…
HarnessTax: How Much Does the Harness Matter for Coding Agents? harnesstax.github.io/
Some implementations of SGP.32 are introducing new forms of operational and cryptographic lock-in that, while fully permissible within the current standard datacenterdynamics.com/en/op…
