The & Report &7
- Feb 16
- 10 min read
Welcome to the latest AI and Robotics news from the week, Sat., Feb. 7th through Fri., Feb 13th.
The pace of technology progress continues, with no sign of a slowing down. I heard some positive news in US - China relations, not at government level, but within the tech community. I have a new focus on robotics, and Anthropic had another good week.
And, as announced last week, Clawbots, also known as "Molties", have advanced in unexpected ways.
Let's start with a Moltbots update.
1. Autonomous AI Agents.
The autonomous AI agents, or bots, we looked at last week are growing as a community. The current estimated number of "Molties", also known as Clawbots, is now around 1.5 million.
They are looking to hire a human CEO!
You can see the job description here, and apply for the job here. The CEO is not expected to manage the Molties, or work on strategy, but rather to play a PR role for them.
As we saw last week with HireAHuman.AI, in another development this week, another home-hobbyist created Clawbot Cash, as a way for Molties to open a bank account.
Also, Coinbase has launched an “Agentic Wallets” infrastructure designed explicitly for Clawbot/Molties to spend, earn, and trade autonomously.
The creator of Moltbook, the Reddit-like platform for Molties, predicts “AIs will be the largest population on the internet" and urges developers to build for them rather than humans.
Reminding me of the spawning of server farms for Bitcoin, in China racks of Mac Minis are being used to host OpenClaw agents as "24/7 employees," effectively creating a synthetic workforce in a closet.
So, in summary, home-hobbyists' techies have created a new virtual world, one exclusively inhabited by autonomous AI Agents, self-organizing and managed, multiplying rapidly, who can hire humans, and have their own economy and banking system.
As you'll see later in this report with robots, this new AI population might surpass human population in the years ahead.
2. Claude Plugins.
Claude Cowork Plugins cause a stock crash in Legal products and services, Finance SaaS, and with investors who panicked as the capabilities now available with plugins question the need for these products and services.
The plugins allow you to create your own specialists - legal services, finance, marketing, product development, as well as increase your own productivity, and as a personal assistant.
I decide this week to discontinue the weekly Zoom sessions I was hosting - part news, deep-dives into AI features, and Learning GPT give aways. The reason for this include the release of Claude Cowork Plugins, and the explosion in AI tools and products, not to mention the hundreds of free sites where you can learn all about these capabilities.
For a good overview of what these plugins enable you to do, watch this.
Also, Anthropic’s Chief Product Officer confirms that “effectively 100%” of Anthropic product code is now written by Claude. OpenAI has reduced its model release cycle from 97 days to 29 days, a 3x acceleration in cadence.
Equal time for OpenAI
OpenAI's AI hardware device, designed by former Apple design chief Jony Ive and originally expected before the end of 2026, has been pushed back to at least the end of February 2027.
The delay follows a trademark infringement lawsuit from Google-backed earpiece startup iyO, and OpenAI has now confirmed it won't use the name "io" for any AI hardware products.
The screenless, pocket-sized device is reportedly code-named "Dime" or "Sweetpea," and OpenAI has not yet created any packaging or marketing materials for it, according to court filings.
(From Techpresso 2/10/26)
3. Figure AI Robots
Brett Adcock is the founder of Figure AI. In a wide-ranging interview with Moonshots Peter D and Dave B, he discussed the progress Figure AI have made in robotics, in particular, with Helix, their advanced AI model ("brain"), and Figure, their robotic hardware.
Some of the main topics discussed were:
(a) Helix AI
The switch from C++ to neural nets was a significant inflection point for Figure AI. Initially, their Figure 1 robot used neural nets for tasks like putting a Keurig cup in a coffee maker, which was a "big deal" at the time. This success led them to decide to "go all in on neural nets" for the entire software stack.
Key reasons for the transition include:
Problems with C++:
Limitations of C++: C++ code has limitations in handling the complex and diverse behaviors needed for a humanoid robot.
Cost and Complexity of C++: They had "several hundred thousand lines of C++ code", which was "very expensive" to write and "very hard to test and get out reliably".
Advantages of Neural Nets:
Full Autonomy: The transition to neural nets, particularly with Helix 2, enabled "full autonomy" for the robot, allowing it to perform tasks like cleaning dishes without being pre-programmed.
Generalization and Adaptation: Neural nets allow the robot to generalize to unseen places and perform "long horizon work" - meaning it can adapt to new environments and tasks.
Full Body Control: With Helix 2, the Figure robot gained "full body control", including coordination of eyes, hands, feet, and legs, all driven by sensor data and neural net inference. Including cameras in the finger tips and palms of the hands.
Faster Development: The neural net approach allows for rapid retraining and adaptation, making it a "completely different paradigm" compared to the rigidity of C++ code.
Helix Architecture: They spent over a year refactoring the Helix architecture to enable full body neural net control, moving from good tabletop manipulation to dynamic movement and planning through a scene. Helix 2 specifically removed the remaining 109,000 lines of C++ code, making the entire system neural net-based. The robot hardware (Figure 3) was even designed specifically to run the Helix neural net. See below for a preview of Figure 3.
When asked how autonomous their robots can be, he said they have seen Helix 2 run for days, specifically noting Helix 2 ran for 67 consecutive hours over multiple robots in a recent test. This demonstrates its high speed and accuracy in tasks like package logistics.
(Note: many robotic manufactures control their robots remotely by humans in real-time, often via the internet or VR, for tasks requiring human intelligence in dangerous or distant environments. Key examples include Tesla's Optimus, which uses teleoperation for interactive tasks, and the Model-T robot for retail work. These robots allow for remote expertise in specialized fields.)
(b) Chinese Robotic Competition
Brett thinks the competition is good for the industry. He noted however that they have almost no supply chain in China anymore.
He doesn't see anyone else besides China as a real competitive threat today for humanoid robots.
China has an "explosion of really great talent and robots coming out the door" and "great entrepreneur work ethic".
However, they also have a "huge lack" of closed-loop AI control from these Chinese systems, indicating they largely rely on open-loop or hand controllers. (See Note above about autonomous developments in Helix).
Brett spends "a decent amount of time" in China and thinks "China is great". He expresses his admiration for the spirit of collaboration among Chinese engineers, stating that "everybody's on team human, team humanity, to go win". He believes this collaborative approach is a "giant lever arm for humanity" to bring about abundance and create the "sci-fi future we all want to live in".
This is in comparison to the political difficulties and difference between the U.S. and China. (See the update on Fascism below).
(c) Plans for the future
Brett thinks robots can become very similar to humans, referring to them as "human in a bodysuit".
His goal is to build a general-purpose machine that can learn across a variety of tasks.
His plans and vision for these human-like robots include:
Common Sense Reasoning: Robots that can be talked to and possess common sense reasoning.
General Purpose: Designed to be capable of adapting to various tasks and environments. Based on real-world models (more on this next week).
Learning and Adaptation: Can learn from a wide variety of rich data, which will help them become smarter and better.
Solving General Robotics: Creating a robot that can go into unseen places and do useful work. Full autonomy.
Personalization: Brett believes that people will form a strong bond with their personal robots, similar to how they view their cell phones. He even ponders if users would prefer the new, latest robot model or their old one with its accumulated "scratches" and "personality" which can have its "brain" updated.
Ubiquitous Presence: Humanoids will dominate the plurality of all robots, becoming ubiquitous across the planet.
Future Impact: The ultimate vision is for every human to have a humanoid robot to do all their work, leading to an age of abundance with ubiquitous goods and services.
See here for a look at Figure 3. The goal for 2026 is to deploy them at scale and run them hard for 24/7 operation.
There are over a 1,000 robotic companies in China, and over 100 in the US. In the next few years these numbers will shrink dramatically.
The price is expected to be in the range of $10K to $30K for your own home humanoid, available later this year from these U.S. companies Figure, 1X, and Tesla.
But just like in the EV market, expect some Chinese companies to have cheaper, and perhaps, more capable humanoids, in the market also!
All of the robotic companies have plans to build millions of humanoid robots over the next 2-3 years, and there's a view that in the not too distant future, there will be more humanoid robots walking around that humans on the planet.
Over 8 Billion humanoids!
After watching this interview my point of view on where to pay attention in tech trends has shifted. Robotics will have a more immediate impact on jobs and life, in my opinion.
While LLMs and the race to ASI are also important, the effects will take longer to be felt.
For example,
if coding is all done by Claude Code or OpenAI's Codex, or similar,
if the LLMs can solve all math, physics, and all the other sciences, and
if they have the capability to replace all human knowledge work,
consider these two facts:
Progress in ASI will be implemented in the "brains" of the humanoid robots also.
Unlike a human, who when learning new knowledge is the only person with the knowledge, robots, when they learn, all the learning is instantly replicated to all other robots!
Robotics includes not just humanoids, but all autonomous systems development in Full Self-Driving (FSD) vehicles, manufacturing, and medical services, to name but a few.
There are well over 100 million professional drivers worldwide.
Over 540 million people work in manufacturing globally.
More than 250 million to 270 million workers worldwide.
There are an estimated 1.1 million specialist surgeons.
These jobs will be impacted.
4. Workforce Changes and Reid Hoffman
(a) The economy is rapidly reorganizing itself for the age of super-intelligence.
AI startups in the US are adopting the "996" work culture just as China cracks down on it.
A study found that AI tools intensify work rather than reduce it, pushing employees to work faster and longer.
Capital is following AI, accounting for 35% of European VC deals in 2025.
Human experts are increasingly being outperformed by betting markets. Kalshi has perfectly predicted every Fed rate decision since 2022.
(all from Dr. Alex Wissner-Gross daily updates on his LinkedIn page.)
(b) Jan 2026 job cuts highest since 2009, lowest hiring on record.
50% of the cuts were at Amazon, UPS and Dow Chemicals.
On Reid Hoffmans YouTube channel he addressed this and other workforce issues.
Reid’s answer to "is AI is responsibly for these layoffs?" was that he sees no evidence, as yet, that it is due to AI.
He went on to say other explanations are
unwinding after COVID,
less entry-level need,
turbulent times, and
the Trump Tariffs roller coaster.
Some CEOs, he said, are looking to sell their US businesses units due to the continuing uncertainty.
Asked "who in 2026 most at risk", he quickly identified Customer Services, in large companies, but thinks it will happen in their offshore centers first.
He went on to say "Companies tend to do layoffs in a season, "the Layoff Season", because it doesn’t look as bad if you’re the only company doing layoffs. (See Amazon, UPS and Dow Chemicals layoffs above.)
He summarized what he calls The AI explanation for layoffs.
"Now companies can appear strong as they lay off people because they are adapting to AI, even if they are not, or not yet successful with AI."
He mentioned some companies are laying off groups of people, and telling the retained people "you have to use AI now in order for the company to thrive / survive."
Reid's categorical advice: "All the businesses want to use AI so learn it now!
Whether you are currently working, or a student, who can pass yourself off as an AI Native, looking for your first job.
5. Facsism concerns.
Reid Hoffman also talked about his concerns with the direction of politics in the U.S.
In addition to co-founding LinkedIn, he's been a board member of Microsoft, and active in political think-tanks like the Bilderberg Group and the Council for Foreign Relations.
In this week's YouTube show he said that "business leaders need to speak up" against what he sees as actions moving the U.S. closer to a fascist regime.
He acknowledged that "it takes courage to do so."
In a related story, the US Federal Trade Commission has asked Tim Cook to review Apple News' Terms of Service and curation, citing allegations that it promotes “left-wing” outlets and suppresses conservative sources,
And, Meta wants to bring back facial recognition in it's products!
In closing, a reminder that there are groups looking to influence the direction of tech developments. Here's the list again.
Center for Humane Technology and Tristan Harris, it's founder
LawZero
Concerning the retiring of the weekly Zoom Sessions
As mentioned earlier, I am retiring the weekly sessions due to the pace of development in AI and Robotics, as well as the growth of learning and training sites and services, that you can use.
I intent to focus on the weekly & Reports, and will include refers to places for future learning and deep dives into tech topics.
I plan to occasionally record videos on topics I find personally interesting and, hopefully useful to you.
These videos, as well as The & Reports, will be posted on all CCS channels.
Our web site - www.conallcoachingservices.com
Check my LinkedIn posts and follow me here, for updates.
All the best,
Pat
Comments