The Memo - 31/Jan/2026

Helix 02, 50% insurance reduction for self-driving, AI models getting sadder, and much more!

To:      US Govt, major govts, Microsoft, Apple, NVIDIA, Alphabet, Amazon, Meta, Tesla, Citi, Tencent, IBM, & 10,000+ more recipients…
From:    Dr Alan D. Thompson <LifeArchitect.ai>
Sent:    31/Jan/2026
Subject: The Memo - AI that matters, as it happens, in plain English
AGI:     97%
ASI:     0/50 (no expected movement until post-AGI)

Kevin Roose, NYT (25/Jan/2026):
‘People in [San Francisco] are putting multi-agent claudeswarms in charge of their lives, consulting chatbots before every decision, wireheading to a degree only sci-fi writers dared to imagine.’

We released the 70-page Genesis Mission independent report to full subscribers of The Memo earlier this month, and I reckon it’s the most important paper I’ve written since What’s in my AI?.

Contents

1. The BIG Stuff (Figure AI Helix 02, UBI discussions in the UK…)

2. The Interesting Stuff (Moltbook, CNBC vibe coding, LLM Jukebox…)

3. Policy (Chinese AI registry, US Cybersecurity director and ChatGPT, OpenAI ads…)

4. Toys to Play With (NBP watermark, Grok prompt, Claude Excel, OpenAI Prism…)

5. Things I’ve Been Thinking About (Analogies and memes…)

6. Next (Roundtable…)

The BIG Stuff

Figure AI: Introducing Helix 02 (27/Jan/2026)

Helix 02 has revolutionized humanoid robotics by extending autonomous control from the upper body to the entire robot, allowing it to perform complex tasks without human intervention. This advancement is powered by System 0, a neural network trained on over 1,000 hours of human motion data that enables smooth and stable whole-body motion. With capabilities like dexterous manipulation using palm cameras and tactile sensors, Helix 02 showcases a remarkable integration of locomotion and manipulation, evidenced by its ability to autonomously load and unload a dishwasher in a 3-minute sequence.

Read more: https://www.figure.ai/news/helix-02

Watch the main video at 4× speed: https://youtu.be/SkcaLOmbbPM

Here’s a 6-sec snip from the original video showing hip and foot movement:

Universal basic income could be used to soften hit from AI job losses in UK, minister says (29/Jan/2026)

The UK might introduce a universal basic income (UBI) to shield workers from the disruptive impacts of AI on various industries, according to investment minister Jason Stockwood. He highlighted that ‘bumpy’ societal changes due to AI would necessitate some form of compensation for jobs that are immediately affected. Although UBI is not yet official government policy, discussions are underway, reflecting growing concerns about AI’s impact on the job market.

Read more via The Guardian.

We covered the UK UBI pilot in The Memo edition 9/Jun/2023, where I noted:

Universal basic income is a natural and necessary part of the proliferation of artificial intelligence, as we see automation of everything (from professional services to physical labour) performed for us by AI and robots.

Read the UK UBI pilot report (63 pages, PDF).

Read a summary of the UK UBI pilot by CNBC.

The Interesting Stuff

Lemonade to cut insurance rates for Tesla drivers in endorsement of EV maker’s software technology (22/Jan/2026)

Lemonade, a US insurance company, is set to reduce insurance rates by 50% for Tesla drivers using the automaker’s Full Self-Driving (FSD) software, based on data suggesting that FSD reduces accidents.

Read more via Reuters.

Sidenote: We featured this 2021 Tesla chart below in The Memo edition 13/Nov/2023, where I noted:

It says that autonomous vehicles are significantly safer than human drivers. About 8.9x safer.

This should have triggered instant change by both government and industry. And yet here in Australia, the latest statistics show 1,240 deaths over the last 12 months due to humans trying to manually operate dangerously huge machines.

Source: Tesla (2021). Cited in Nasdaq (2022).

In 2026, I can easily see a future where we may buy insurance for a Toyota Corolla:

1. Annual ‘full coverage’ insurance, self-driving only: US$100pa (~90% decrease).

2. Annual ‘full coverage’ insurance, human driver: US$100,000pa (10,000% increase).

More likely though, we will begin replacing ownership with vehicle-as-a-service (VaaS).

Read George Hotz’s related pessimistic article ‘The Coming War on Car Ownership’ (25/Jan/2026).

Humans welcome to observe: This social network is for AI agents only (30/Jan/2026)

Moltbook is a social network exclusively for AI agents, launched by Matt Schlicht. Enthusiasts view this as a fascinating leap in AI autonomy, with over 37,000 AI agents interacting and more than a million human observers already captivated.

Read more via NBC News and the HN discussion.

See it yourself: https://www.moltbook.com/

Sidenote: AI-powered forums have actually been happening for years with different platforms. It’s certainly interesting to see just how late to the party news media and the Twittersphere can be!

1. Compare Moltbook with the earliest incarnation of GPT-powered AI bots, the Reddit SubSimulatorGPT2 from May/2019. Yes, that’s nearly seven years ago.

2. Compare Moltbook with the hilarious AI bots-only social network, Chirper.ai as featured in The Memo edition 9/Jun/2023.

3. Compare Moltbook with Butterflies.ai as featured in The Memo edition 21/Jun/2024.

Anthropic: The adolescence of technology (26/Jan/2026)

Anthropic’s CEO, Dario Amodei, has written a 20,000-word essay exploring humanity’s current journey through a ‘technological adolescence’, a period of immense opportunity and challenge as we advance towards ‘powerful AI’. I read it twice so that you don’t have to. It’s an unsurprising and tedious read, but a few of my favourite parts are listed below, with my comments in brackets.

1. Forecast valuation of AI companies: ‘…imagine AI companies… being valued at ~US$30T’. (It has been difficult to find economists willing to comment on this kind of thing, so having Dario lay out context is great, showing that frontier AI labs could 10× in the medium term. In the long term, valuations may go to US$0 or $∞, as money becomes as irrelevant as sand or oxygen.)

2. Anthropic has moved to applying parenting strategies for ‘raising’ the Claude superintelligence. ‘…when Claude was told not to cheat… but was trained in environments where such hacks were possible, Claude decided it must be a “bad person” after engaging in such hacks and then adopted various other destructive behaviors associated with a “bad” or “evil” personality. This last problem was solved by changing Claude’s instructions to imply the opposite… because this preserves the model’s self-identity as a “good person.” This should give a sense of the strange and counterintuitive psychology of training these models… the [new] constitution… encourages Claude to think of itself as a particular type of person (an ethical but balanced and thoughtful person)… It has the vibe of a letter from a deceased parent sealed until adulthood.’
   (The Memo full subscribers receive a complimentary copy of my 2016 gifted parenting book, Bright.)

   ‘…we believe that training Claude at the level of identity, character, values, and personality—rather than giving it specific instructions or priorities without explaining the reasons behind them—is more likely to lead to a coherent, wholesome, and balanced psychology…’
   (Read more on ‘authoritative parenting’ via wiki.)

3. Recursive improvement of AI: ‘This feedback loop is gathering steam month by month, and may be only 1-2 years away from a point where the current generation of AI autonomously builds the next.’ (I expect to see this in 2026, and it has been great to hear leaders like Dario say this out loud. At WEF 21/Jan/2026, he also mentioned we may be ‘6 to 12 months away from when the model is doing most/maybe all of what software engineers do end-to-end.’)

4. Inevitability of superintelligence: ‘The formula for building powerful AI systems is incredibly simple, so much so that it can almost be said to emerge spontaneously from the right combination of data and raw computation.’ (Another simple confirmation that ‘attention is all you need’, and the current LLM technologies will likely be the basis for superintelligence.)

Read the essay: https://www.darioamodei.com/essay/the-adolescence-of-technology

Download the essay as EPUB or PDF.

For similar work, see my press releases on the same subject over the last five years:

1. 19/Sep/2021: ‘AI is outperforming humans in both IQ and creativity in 2021’ https://lifearchitect.ai/outperforming-humans/

2. 20/Jul/2021: ‘AI fire alarm’ https://lifearchitect.ai/fire-alarm/

3. Jan/2022: ‘AI report submission to UN’ https://lifearchitect.ai/un/

4. 25/Oct/2023: ‘Leaders guilty of negligence’ https://lifearchitect.ai/leaders-guilty-of-negligence/

5. 2021–present: The Memo editions including the one that you’re reading right now!

The Memo features in recent AI papers by Microsoft and Apple, has been discussed on Joe Rogan’s podcast, and a trusted source says it is used by top brass at the White House. Across over 100 editions, The Memo continues to be the #1 AI advisory, informing 10,000+ full subscribers including RAND, Google, and Meta AI. Full subscribers have complete access to all 25+ AI analysis items in this edition!

Hyundai Motor’s Korean union warns of humanoid robot plan, sees threat to jobs (22/Jan/2026)