HOW TAIWAN RELAUNCHED AI FOR HUMAN BRAIN DEVELOPMENT - 1987, 1993, 2001-2, 2004 2009, 2012 +++
In 1956 Von Neumann's last notes were "Computer and Brain". While Neumann-Einstein-Turing world to transform humanity in line with nature's deep data science goes back to 1905 Einstein e=mcsquared, Neumann and Turing had little more than 10 post war years to design computers and data coding. Neuman anticipate neural network algorithms and at least 10**18 more tech being expoentialised by 2025 through computers chips and satellites. He advised design win-win trades. End making energy engineers nations most hated and wanted of people.
Taiwan became the epicente of what human brains could do with machine learning by betting the country on chips - see counyry's tech grandfather H Li and eg leaders of Foxconn world number 1 in advnaced manufactring conumer products and Maurice Changs ai foundry from 1987. Taiwean was also in the middle of diaspora chinese investments made mainly through british-influenced chiese fianacial centers hong knog (see eg how Li Ka Shing helped continue investment after HK return toi china) see eg intelligence griwn around singapore liee kuan yew.
Jensen Huang started nvidia 1993 to connect taiwan and american intelligence and design computing 2.0 around deep data and gaming's non linerar processing. 1993 was a good time - as moores law's accelerator negun in 1965 was waning; the world needed to choose how data and messages linked humans on worldwide webs.
Here are some dates when emerging tech multiplied ariund seeds taiwan-valley sewed - which are AI and which AI depended on are up to you
by 1997 3 taiwan american families -yang's yahoo, tsias, as well as huangs nvidia were uniting best of pacific coast's valley-taiwan (and hk, and japan , and soon with jack ma china). This shouldnt surprise teh wordlwide web semed an extroardinary brain freedom in its first few years before advertsising media marooned it in not-always socieal good media
in 2001 steve jobs hosted 65th borthday party with fazle abed- he wanted to revisit his 1984 launch of ending big brother with the g=former shell oil engineer who had dedicated since 1972 to village women empowerment - why hadnt bangladesh shared its intelligence with world- abed said there had been a lot to do but he'd start a university iif silicon valley would start celebrating femeale engineers and opening their work to female graduates the world over; by 2002 jobs and jensen huang had started coding pixels; by 2004 y=tje yangs were invetsing in an interdiciplianr as well as female and male lac at stanfird;s engineering quad- this gave space for such women as koller founder pof moocs, fei-fei li founder of machine learning, lila inbrahim to become demis hassabis ops leader out of london, to perco=ieve human senses, as well as women The Economist first met 1982 in why not silicon valey everywhere such as mrs john doerr and condi Rice and over in berleley the soon to be nobel laureatate jennifer doudna of CRiSPR gene editing. It was 2009 huang bet the compoany on ai also starting deep learning lab at stanfird quad; it was 2012 venture capital piled in with alexnet winning fei-fei li's annual competition , it was 2015 elon musk became first co=ustomer for the biggest chip brains ever made-all the while it was the peopels of taiwan that were manufatiring the products ai plays on
To understand how taiwan has played at least as much genius-design on nvidia's innovations as silicon valley - see nvidia annual summits both from santa clara and from taiwan. In 2025, taiwan is one of first 10 ecosystem epicenters of nvidia's 25X energy more efficient supercomputers and links to nvidi'as worldwide inception catalogue of 20000 ai startups-
survey what taiwan hosts on suoercomputer - eg earth2.0 aiming to be locally most advanced weather alerts as well as all the deepest data of climate mapping. At the very least it is vital that peoples have these open maps not just insuraince companies. Another lead out of taian is digital twinning. It is important to note transparency and deepth of partner formation around a project such as earth2.0 Our understanding is 3 first partners were nvidia, taiwan government and teh weather companu which had been bought by IBM. IBM devotion to long term daya collection on weather is important as it is currently also a leader in quantum compiuting . It is likely that the depth of data mapping weathr calls for will one day need quantum's best. NB while nasa has pioneered spce work, it is not clear that the risks from space through 21sdt centiry is eomething american taxpayer want to be sole invetsor in - at least that seems to be demonstarted by the markets for space rocket launches.
I am not an expert on space exploration markets- here is an introductiry view from Grok3 summer 2025- I expect this may be fast chnaging and that there may be other ways of defining what sort of space race deepens either a compay's sovereignty or an nation's depending on how deep or 360 degree broad its interest in space monitoring is> However what we can be certain of - suorecompute gets connected with accurate but contimous mapping space involves if it is to be safely united for all peoples.
- Rocket Launching (Space Transportation)
- In-Space Manufacturing
- Space Agriculture
- Nuclear Propulsion
- Satellite Services/Maintenance
- Space Exploration/Human Spaceflight
- Space Tourism/Habitats
- Companies:
- SpaceX (USA): ~87% global upmass (2023), 96 launches, reusable Falcon 9 ($2,000/kg) [web:12, web:20].
- Rocket Lab (USA/NZ): Small-lift leader, 50+ Electron launches, Neutron in dev [web:14, web:19].
- ULA (USA): Vulcan for NASA/DOD, 100+ missions since 2006 [web:8, web:22].
- Arianespace (France/EU): Ariane 6 debut 2024, historic 50% market share [web:9, web:13].
- Governments:
- USA (NASA, DOD): SLS, EELV, $6.8B private contracts [web:5, web:10].
- Russia (Roscosmos): 18 launches (2023), Soyuz for ISS [web:12].
- China (CNSA): 60+ launches, Long March for Tiangong [web:12, web:20].
- India (ISRO): 7 launches, cost-effective PSLV ($15,000/kg) [web:12].
These days factories which blend human and robot inputs are most smartly designed with a virtual twin accompanying the first real factory. That way safety can be simulated; and flexible redesign of interactions between humans and robots achieved. The robots collect all the data. This also makes replicating smart factory franchises to other nations practical. For example foxconn;s digtal twins will likely set the benchamrk as will hong kongs where the world most affordable blended medica cenetre is being built as announced at HKUST 2024
We would be surprised if taiwan cities are the first to sewe humanoids out populating humns doing all the repetitive work of city cleaning etc. Please note agentic ai without robot bodies will come first beuase physical ai needs a lot of repetitive learnning to be added. Taiwan is already nearly to in high school maths and it will be no surprise if its first to transform school maximising role ai agents can augment both teachers and stydent roles.
