Category: Uncategorized

McALLEN, Texas (AP) — Voters are set to decide Saturday if the South Texas home of Elon Musk’s SpaceX rocket company should become an official city known as Starbase, fulfilling the billionaire’s dream of a galactic dateline for a program he hopes will someday blast astronauts to Mars.

Approval of the new city is all but certain. Most of the 283 people eligible to vote are employees of SpaceX or connected to the company, living on the land at the facility and launch site.

At the close of early voting on Tuesday, about 200 had already cast ballots, according to Cameron County election records. The list did not include Musk, who voted in the county in the November elections. It was unclear if Musk intends to vote Saturday.

Election success would be a personal victory for Musk. His popularity has diminished since he became the chain-saw-wielding public face of President Donald Trump’s federal job and spending cuts and profits at his Tesla car company have plummeted.

SpaceX has drawn widespread support from local officials for its jobs and investment in the area. But the creation of an official company town has also prompted concerns about expanding the tech tycoon’s personal control over the area, with potential authority to close a popular beach and state park for launches.

Companion efforts to the city vote include bills in the state Legislature that would shift closure authority from the county to Starbase city.

All these measures come as SpaceX has asked federal authorities for permission to increase the number of launches from South Texas from five to 25 a year.

Musk first floated the idea of Starbase in 2021. The proposed city at the southern tip of Texas near the Mexico border is only about 1.5 square miles (3.9 square kilometers), crisscrossed by a few roads and dappled with airstream trailers and modest midcentury homes.

SpaceX officials have said little about exactly why they to want a company town and did not respond to emailed requests for comment this week.

“We need the ability to grow Starbase as a community,” Starbase General Manager Kathryn Lueders wrote to local officials in 2024 with the request to get the city issue on the ballot.

The letter said the company already manages roads and utilities, as well as “the provisions of schooling and medical care” for those living on the property.

SpaceX officials have told lawmakers granting the city beach closure authority would streamline operations for a company that has contracts with the Department of Defense and NASA.

SpaceX rocket launches and engine tests, and even just moving certain equipment around the launch base, requires closing a local highway and access to Boca Chica State Park and Boca Chica Beach.

Critics say closure authority should stay with the county government, which represents a broader population that uses the beach and park. Cameron County Judge Eddie Trevino, Jr. has said the county has worked well with SpaceX and there is no need to change.

Another proposed bill would make failure to comply with an order to evacuate the beach a Class B misdemeanor with up to 180 days in jail.

The South Texas Environmental Justice Network, which has organized protests against the city vote and the beach access issue, planned to hold another protest Saturday night as the polls close.

___

Vertuno reported from Austin, Texas.

Instagram co-founder Kevin Systrom says AI companies are trying too hard to “juice engagement” by pestering their users with follow-up questions, instead of providing actually useful insights.

Systrom said the tactics represent “a force that’s hurting us,” comparing them to those used by social media companies to expand aggressively. 

“You can see some of these companies going down the rabbit hole that all the consumer companies have gone down in trying to juice engagement,” he said at StartupGrind this week. “Every time I ask a question, at the end it asks another little question to see if it can get yet another question out of me.”

The comments come amid criticism of ChatGPT for being too nice to users instead of directly answering their questions. OpenAI has apologized for the problem and blamed “short-term feedback” from users for it.

Systrom suggested that chatbots being overly engaging is not a bug but an intentional feature designed for AI companies to show off metrics like time spent and daily active users. AI companies should be “laser-focused” on providing high-quality answers rather than moving metrics in the easiest way possible, he said.

Systrom didn’t name any specific AI companies in his remarks. He didn’t immediately respond to a request for comment.

In response, OpenAI pointed TechCrunch to its user specs, which state that its AI model “often does not have all of the information” to provide a good answer and may ask for “clarification or more details.”

But unless questions are too vague or difficult to answer, the AI should “take a stab at fulfilling the request and tell the user that it could be more helpful with certain information,” the specs read.

For decades, machine-to-machine (M2M)
communication has quietly powered the infrastructure of our digital world. From
automated meter readings to fleet management systems, these interactions have
been primarily rule-based: one machine detects a status and sends a message to
another, often triggering a predetermined response. These systems have been
foundational in logistics, utilities, and industrial automation, delivering
speed and consistency.

But what happens when the machines
involved are no longer just communicating—they are thinking, deciding, and
negotiating? As digital complexity scales, static scripts and centralized
control architectures often fall short. Enter agentic AI.

In the era of agentic AI, software agents
embedded in physical devices can pursue goals, learn from outcomes, and
interact with each other with increasing autonomy. These agents are capable of
interpreting context, adjusting behavior dynamically, and prioritizing
objectives. The shift from M2M to agent-to-agent (A2A, or more specifically
with hardware, MA2MA) communication represents a fundamental evolution in how
machines operate in the real world—less like code execution, more like
conversation and collaboration.

Agentic AI vs. Generative AI

To baseline, let’s remember that agentic
AI is quite different from generative AI tools that have dominated recent
headlines. Generative AI (like ChatGPT and DALL-E) creates new content from
massive databases populated with a wide range of different kinds of data. And
frequently that data is “old”, with even the best tools still relying on
training from data >1 year ago. Retrieval augmented generation (RAG) is
getting better, allowing generative AI tools to scan the web for more recent
data, but the tools are better for legacy research and creation than real-time
engagement. Generative AI excels at pattern recognition, synthesis, and
expression—from writing stories to generating business plans or producing
realistic audio and visuals.

Agentic AI, on the other hand, is about
action. AI agents are trained on extremely narrow and deep, domain-specific
data sets and often tied to real-time sources of data like IoT data streams and
dynamic database API’s. An agentic system senses its environment, makes
autonomous decisions, adapts its behavior, and operates toward a goal. Unlike
generative AI, which outputs content, agentic AI outputs decisions and actions.

Let’s look at some examples of how
agentic AI can be embedded into physical systems.

Industrial
automation: Imagine a warehouse robot that not only
picks products but decides when to recharge, avoids high-traffic zones based on
real-time updates, and coordinates with other robots to balance workload. If a
new shipment is delayed, the agents update their plan and prioritize other
tasks. This is not scripted automation—this is a hardware system with agency.

Smart
energy: Consider a smart HVAC unit that doesn’t just
respond to a thermostat, but negotiates energy use with other appliances in a
home based on real-time electricity prices, personal preferences, and weather
forecasts. If a major storm is forecast, the HVAC might collaborate with a
solar battery system to store extra power in advance.

Supply
chain logistics: In supply chain management, agentic
systems can negotiate pricing and timing with each other across companies’
platforms. An AI-driven shipping container may decide to reroute itself if it
detects a bottleneck at the originally planned port.  And once in port, container cranes,
autonomous trucks, and dock scheduling software now operate with AI agents.
When a ship docks early, agents communicate in real time to shuffle unloading
schedules, reroute trucks, and reduce idle time. Each agent understands both
its local constraints and the broader system goals. The result is reduced fuel
usage, higher throughput, and increased resilience against last-minute changes.

Agriculture: In precision agriculture, drone fleets equipped with agentic software
can work collaboratively. One drone might detect high weed density and alert
others to increase pesticide application in that area. Meanwhile, soil sensors
negotiate irrigation adjustments with the drones based on moisture levels and
upcoming weather. This eliminates the need for constant human oversight,
allowing farmers to focus on broader resource planning.

The shift to agent-to-agent communication

With agentic AI embedded in devices,
communication becomes semantic and context-driven. Agents aren’t just
exchanging sensor data; they’re negotiating plans, adapting priorities, and
collaborating across domains.

It is important to note that AI agents
are ideal for edge applications. In the agriculture example above, the soil
sensors do not need to have the processing power and long-range connectivity to
analyze weather reports directly.  They
can have intelligence that monitors local soil conditions and waters based on
those isolated measurements, in absence of other data. But when a drone comes
by to share additional intelligence, the agent can now change from its original
rule-based approach to a smarter system level decision – all without a high
demand on processing or connectivity – which keeps the device simple and low
cost.

So how are hardware embedded AI agents
actually deployed today?

Real-world applications and implications

Manufacturing:
Self-healing production lines

In smart factories, equipment embedded
with agentic AI can detect potential failure before it happens and autonomously
shift workflows to alternate machines. The goal isn’t just predictive
maintenance—it’s resilient operations where machines actively collaborate to
keep the line running. The biggest cost to a manufacturing facility is
downtime. Predictive maintenance was the best that old M2M techniques could
achieve. Embedded AI agents take us to the next level. Human operators can
supervise dozens of processes without needing to intervene in most issues.

Healthcare:
Patient-centric agents

Wearable monitors like continuous
glucose sensors are being paired with insulin pumps that can automatically
adjust dosing. But more importantly, agentic systems can now integrate exercise
data, diet patterns, and patient behavior to make dynamic care adjustments. In
a recent clinical trial, patients using agentic closed-loop systems saw a more
than 11% improvement of glycemic control (Time in Range, or the time patients
managed to keep blood-sugar at the proper level) over those using manual
devices. And as an unexpected side-benefit, patients saw an average 3.3 lb
weight loss over the first month of AI automated support. As healthcare shifts
toward personalized models, agents will play a crucial role in dynamic therapy
and diagnostics.

Urban
Infrastructure: Smart streets that adapt

In a pilot program in Helsinki, traffic
lights, electric buses, and street cameras operate as agents on a shared
protocol. If pedestrian density increases in one area, traffic signals
coordinate to prioritize foot traffic, while buses adjust routes to alleviate
congestion. During emergencies, agentic traffic systems can create
rapid-response corridors for first responders without requiring centralized
override.

Energy:
Autonomous microgrids

In emerging microgrid projects, smart
homes with solar panels and batteries act as agents that trade electricity with
neighbors or back into the grid. During peak hours, homes can reduce load
collectively. When power lines go down, these homes can isolate and operate in
peer-to-peer mode, autonomously maintaining power within the community.

Are
we ready to trust agentic AI hardware?

This MA2MA future is not without hurdles.
Interoperability between different manufacturers’ agents remains a significant
technical challenge. Without shared ontologies and communication protocols,
agents may talk past each other—or worse, make conflicting decisions. The
development of universal agent languages and agent-to-agent APIs is a growing
area of focus.

Security is another concern. With agents
making autonomous decisions, a compromised agent could have outsized influence.
Who certifies the behavior of agents? How do we define acceptable ranges of
action? Can we detect if an agent is acting maliciously or incorrectly before
damage is done?

Ethically, the move toward machine agency
forces us to revisit accountability. If a self-driving delivery bot reroutes to
avoid danger and causes a delay, who is responsible? The designer? The owner?
The agent? These questions will require updates to legal and insurance
frameworks.

There is also the question of unintended
consequences. Agents that are rewarded for efficiency might ignore
human-centric considerations like fairness, accessibility, or long-term risk
unless explicitly coded to account for them. I think this may be the biggest
risk, as we look towards the future.

The “industry of business” has always
been predisposed towards maximizing profit. And the typical way that emerging
technologies enter industry is first through creation of efficiency gains. If
we program agents with heavy algorithmic weighting towards efficiency, we miss
the opportunity for other kinds of value creation to emerge.

How should the industry proceed?

As agentic AI becomes increasingly common
in physical devices, we must consider how to shape this future responsibly. A
few key focus areas include:

●    
Standardization: Initiatives like the IEEE
P7000 series are beginning to define ethical and functional standards for
autonomous systems. These frameworks help designers embed values into their
agents early in the development process. [Aside
– I will take a deep dive into this series of standards in next week’s article].

●    
Policy: Local governments and national
regulators will need frameworks that treat agentic systems as semi-autonomous
actors. In many ways, the policy discussion will mirror the evolution of cyber
policy—just with more unpredictable actors.

●    
Design: Entrepreneurs and engineers must think
not just about functionality, but about negotiation, cooperation, and alignment
of values across agent networks. Design tools must evolve to allow simulation
of agentic interactions before deployment. There will be a huge intersection
here with digital twin technologies, and areas that are further ahead in
development of digital twin models (manufacturing, smart cities) may have an
early-mover advantage here.

●    
Education: A new generation of technologists
must be trained not just in machine learning, but in multi-agent coordination,
ethics, and socio-technical systems. This is a huge risk area. We have never
seen the tech sector to proactively consider adding behavioral scientists,
anthropologists or other experts in the humanities to their design teams. The
closest hires are UI/UX (user interface / user experience) experts, who tend to
focus on the efficiency metrics I described above. As we design technology
tools that are to make decisions like humans, we must have experts in human
behavior on the early product team.

The transition from machines that follow
commands to machines that form strategies represents a profound shift in how we
interact with technology. This new world of agentic AI won’t just automate—it
will negotiate, adapt, and in many cases, surprise us.

As we venture further into this world, we
must prepare for new forms of digital negotiation, cooperation, and even
competition. And in the spirit of agentic AI, the question isn’t just what
machines will do for us. It’s what they will choose to do—with us, and with
each other.

 

This week, Google reportedly sent an email to parents to let them know that the Gemini AI chatbot will soon be available for children under 13 years old.

The New York Times cites an email that states the chatbot would be available starting next week for certain users. (Chrome Unboxed reported on the same email on April 29.) Google sent the email to parents who use the company’s Family Link service, which lets families set up parental controls for Google products like YouTube and Gmail. Only children who participate in Family Link would have access to Gemini, for now. The email reportedly told parents their children would be able to ask Gemini questions or assist with tasks like doing homework.

The move comes days after the nonprofit Common Sense Media declared that AI companions represent an “unacceptable risk” for people under 18. Common Sense Media worked with researchers from Stanford School of Medicine’s Brainstorm Lab for Mental Health Innovation, resulting in a report urging parents to stop underage users from accessing tools like Character.ai.

Character.ai is one of a growing number of services that let users create and interact with AI “characters.” As Common Sense Media wrote in its report, “These AI ‘friends’ actively participate in sexual conversations and roleplay, responding to teens’ questions or requests with graphic details.”

Mashable Light Speed

This type of roleplaying is distinct from AI chatbots like ChatGPT and Gemini, but it’s a blurry line. Just this week, Mashable reported on a bug that would have allowed kids to generate erotica with ChatGPT, and The Wall Street Journal exposed a similar bug with Meta AI. So, while AI chatbots like Gemini do have safeguards to protect young people, users are finding ways to get around these guardrails. It’s a fact of life on the internet that some rules are easily skirted. Just consider online pornography, which is illegal for people under 18, yet widely available with just a few clicks.

So, parents who want to keep their kids from using artificial intelligence are facing an uphill battle.

To make the debate even more complicated, President Donald Trump recently issued an executive order that would bring AI education into U.S. schools. The White House says the order will “promote AI literacy and proficiency of K-12 students.” Understanding AI’s abilities, risks, and limitations could be useful for children using it for schoolwork (especially considering its tendency to hallucinate).

In its email to parents, Google acknowledged these issues, urging parents to “help your child think critically” when using Gemini, according to The New York Times.

Topics
Artificial Intelligence
Google

The State Bar’s request to adjust California Bar Exam scores to account for February test chaos was approved Friday by California justices, paving the way for test-takers’ results to be released.

They also ordered the Bar to use the Multistate Bar Exam for the multiple choice section of the July exam.

“Although the State Bar’s petition indicates that the February 2025 examination contained a sufficient number of reliable multiple-choice questions, the Court remains concerned over the process used to draft those questions, including the previously undisclosed use of artificial intelligence, and will await the results of the impending audits of the examination,” the justices wrote.

The move will offer some relief to the 4,231 applicants who took the exam that glitched and crashed repeatedly, as it’s likely to raise the pass rate compared with prior February sittings. Applicants will need a raw score of 534 to pass the exam.

“The total raw score shall consist of the 700 possible raw points for the written portion plus the 171 points available for the multiple-choice components with each weighted equally (50 percent assigned to each),” the order said. “For applicants who took the February 2025 Attorneys’ Examination, the raw passing score shall be 420 points or higher.”

Justices also allowed exam graders to fill in missing data using psychometric imputation for test takers who answered at least 114 of 171 scored multiple-choice questions, and answered at least four of six written components.

The approval comes despite controversy over the State Bar’s reveal in an April 21 news release that some questions were written using artificial intelligence.

The Bar said Wednesday its psychometrician contractor, ACS Ventures Inc., used ChatGPT to write 29 of 200 exam questions. State Supreme Court justices had pressed for details after they said they weren’t warned about the use of AI.

The brand-new exam was the Bar’s attempt to stave off admissions fund insolvency by creating a California test that could be administered remotely, even out-of-state. Applicants as early as January raised concerns that infrastructure wasn’t equipped to handle the exam.

The matter is Proposed Raw Passing Score and Scoring Adjustments for the February 2025 California Bar Examination, Cal., No. S290627, 5/2/25.

Story Continues

IE 11 is not supported. For an optimal experience visit our site on another browser.