MAROKO133 Hot ai: AWS goes beyond prompt-level safety with automated reasoning in AgentCor

📌 MAROKO133 Hot ai: AWS goes beyond prompt-level safety with automated reasoning i

AWS is leveraging automated reasoning, which uses math-based verification, to build out new capabilities in its Amazon Bedrock AgentCore platform as the company digs deeper into the agentic AI ecosystem. 

Announced during its annual re: Invent conference in Las Vegas, AWS is adding three new capabilities to AgentCore: "policy," "evaluations" and "episodic memory." The new features aim to give enterprises more control over agent behavior and performance. 

AWS also revealed what it calls “a new class of agents," or "frontier agents," that are autonomous, scalable and independent. 

Swami Sivasubramanian, AWS VP for Agentic AI, told VentureBeat that many of AWS’s new features represent a shift in who becomes a builder. 

“We are actually on the cusp of a major tectonic transformation with AI, but agentic AI is truly starting to transform what is the art of the possible, and it is going to make this one of the most truly transforming technologies,” Sivasubramanian said. 

Policy agents

The new policy capability helps enterprises reinforce guidelines even after the agent has already reasoned its response. 

AWS VP for AgentCore David Richardson told VentureBeat that the policy tool sits between the agent and the tools it calls, rather than being baked into the agent, as fine-tuning often is. The idea is to prevent an agent from violating enterprise rules and redirect it to re-evaluate its reasoning. 

Richardson gave the example of a customer service agent: A company would write a policy stating that the agent can grant a refund of up to $100, but for anything higher, the agent would need to bounce the customer to a human. He noted that it remains easy to subvert an agent's reasoning loop through, for instance, prompt injection or poisoned data, leading agents to ignore guardrails. 

“There are always these prompt injection attacks where people try to subvert the reasoning of the agent to get the agent to do things it shouldn’t do,” Richardson said. “That’s why we implemented the policy outside of the agent, and it works using the automated reasoning capabilities that we’ve spent years building up to help customer define their capabilities.”

AWS unveiled Automated Reasoning Checks on Bedrock at last year’s re: Invent. These use neurosymbolic AI, or math-based validation, to prove correctness. The tool applies mathematical proofs to models to confirm that it hasn’t hallucinated. AWS has been leaning heavily into neurosymbolic AI and automated reasoning, pushing for enterprise-grade security and safety in ways that differ from other AI model providers.

Episodic memories and evaluations

The two other new updates to AgentCore, "evaluations" and "episodic memory," also give enterprises a better view of agent performance and give agents episodic memory.

An enhancement of AgentCore memory, episodic memory refers to knowledge that agents tap into only occasionally, unlike longer-running preferences, which they have to refer back to constantly. Context window limits hamper some agents, so they sometimes forget information or conversations they haven’t tapped into for a while. 

“The idea is to help capture information that a user really would wish the agent remembered when they came back," said Richardson. "For example, 'what is their preferred seat on an airplane for family trips?' Or 'what is the sort of price range they're looking for?'"

Episodic memory differs from the previously shipped AgentCore memory because, instead of relying on maintaining short- and long-term memory, agents built on AgentCore can recall certain information based on triggers. This can eliminate the need for custom instructions.

With AgentCore evaluations, organizations can use 13 pre-built evaluators or write their own. Developers can set alerts to warn them if agents begin to fail quality monitoring.

Frontier agents

But perhaps AWS's strongest push into enterprise agentic AI is the release of frontier agents, or fully automated and independent agents that the company says can act as teammates with little direction. 

The concept is similar, if not identical, to those of more asynchronous agents from competitors like Google and OpenAI. However, AWS seems to be releasing more than just autonomous coding agents. 

Sivasubramanian called them a "new class" of agents, "not only a step function change in what you can do today; they move from assisting with individual tasks to complex projects."

The first is Kiro, an autonomous coding agent that has been in public preview since July. At the time, Kiro was billed as an alternative to vibe coding platforms like OpenAI’s Codex or Windsurf. Similar to Codex and Google’s myriad asynchronous coding agents, including Jules, Kiro can code, undertake reviews, fix bugs independently and determine the tasks it needs to accomplish. 

AWS security agent, meanwhile, embeds deep security expertise into applications from the start. The company said in a press release that users “define security standards once and AWS security agent automatically validates them across your applications during its review — helping teams address the risks that matter to their business, not generic checklists.”

The AWS DevOps agent will help developers, especially those on call, proactively find system breaks or bugs. It can respond to incidents using its knowledge of the application or service. It also acknowledges the relationships between the application and the tools it taps, such as Amazon CloudWatch, Datadog and Splunk, to trace the root cause of the issue. 

Enterprises are interested in deploying agents and, eventually, bringing more autonomous agents into their workflows. And, while companies like AWS continue to bolster these agents with security and control, organizations are slowly figuring out how to connect them all. 

🔗 Sumber: venturebeat.com

📌 MAROKO133 Breaking ai: Understanding why America’s biggest solar thermal project

One of the most ambitious solar projects in history is quietly heading for shutdown after just a decade of operation. The Ivanpah Solar Power Facility in California’s Mojave Desert was once hailed as a symbol of America’s clean energy future. A $2.2 billion, utility-scale solar thermal plant that promised to power 140,000 homes and prove that big, futuristic renewable projects could work.

Instead, Ivanpah has become a cautionary example about timing, technology bets, politics, and the unforgiving realities of engineering at scale. Its closure is not the end of solar power, but it does show how quickly an industry can change, and how even bold ideas can be overtaken by economics.

A child of the post-crisis green stimulus

To understand Ivanpah, you have to go back to the late 2000s. The US was reeling from the 2008 financial crisis. Unemployment was high, the housing bubble had burst, and the Obama administration was under pressure to revive the economy while tackling climate change.

The 2009 American Recovery and Reinvestment Act poured billions into clean energy as part of a historic stimulus. The idea was simple. Create jobs, cut emissions, and spur a new generation of green infrastructure.

Ivanpah emerged directly from this wave of optimism. BrightSource Energy, led by CEO John Willard, was a pioneer in concentrated solar thermal technology. NRG Energy, headed by CEO David Crane, saw Ivanpah as a bold bet that could reposition the company as a leader in renewables. Google invested $168 million.

Together, they pitched a 392-megawatt solar thermal facility that would generate around 1 million megawatt-hours of electricity a year. Backed by a $1.6 billion federal loan guarantee, Ivanpah was framed as both an engineering marvel and a political statement. Proof that America could build big, clean, high-tech energy projects. Construction began in 2010, and thousands of workers poured into the Mojave Desert to bring it to life.

How Ivanpah’s concentrated solar power worked

Ivanpah didn’t use the familiar flat solar panels you see on rooftops. Instead, it relied on concentrated solar power (CSP), a heat-based technology.

More than 300,000 computer-controlled mirrors, called heliostats, were spread across thousands of acres. These mirrors tracked the sun throughout the day and focused sunlight onto receivers atop three 350-meter-tall towers. The concentrated light heated water into steam, which then drove turbines to generate electricity, much like a coal or nuclear plant, but without burning fuel.

On paper, CSP offered several advantages. It could scale to utility size, promised more stable output than earlier solar technologies, and Ivanpah even had a natural gas backup system to generate power when the sun wasn’t shining. For utilities such as PG&E and Southern California Edison, this was sold as a steady, zero-carbon resource under long-term contracts. In theory, it was elegant. In practice, the world moved faster than the project could keep pace.

A changing market undercut the business case

Ivanpah’s core problem wasn’t just technical. It was economic timing. While the plant was being built between 2010 and 2014, the entire solar industry went through a transformation. The cost of photovoltaic (PV) solar panels fell by nearly 80%, driven largely by massive manufacturing capacity in China. Suddenly, simple PV farms, both rooftop and utility-scale, became cheaper, faster, and easier to deploy than complex CSP systems.

At the same time, the US shale boom sent natural gas prices plunging. Gas-fired power plants became some of the cheapest sources of electricity to build and operate. By the time Ivanpah went online in 2014 with great fanfare, it was entering a completely different market from the one in which it had been conceived. It was a slow, capital-intensive project built for a world where renewables were expensive and needed heavy support. Instead, it arrived in a world where PV and gas were already undercutting it on cost.

To make matters worse, Ivanpah struggled to hit its performance targets. In its first year, it generated only about two-thirds of its promised output. That shortfall frustrated its utility customers and handed critics an easy talking point. Why support a billion-dollar solar plant that couldn’t deliver what it promised?

For David Crane and NRG, the project became a high-risk bet that cooled investor enthusiasm. The vision was bold, but the timing was brutal.

Politics, birds, and the desert tortoise

With so much public money involved, Ivanpah quickly became a political lightning rod. Supporters framed it as a visionary step toward clean energy independence. Opponents labeled it an overpriced “green boondoggle” propped up by taxpayers, pointing repeatedly to the $1.6 billion federal loan guarantee. Some lawmakers questioned why California needed such experimental technology when cheaper options existed.

Unexpectedly, Ivanpah also faced opposition from parts of the environmental community. The Sierra Club, normally a strong supporter of renewables, raised concerns about the project’s impact on the threatened desert tortoise. Biologists warned that the vast field of mirrors would disrupt fragile desert habitats.

Then came the headline-grabbing bird deaths. Birds flying through the intense beams of reflected sunlight were reportedly killed, some igniting midair and creating “streamers” seen by plant workers. What had been marketed as a climate solution was suddenly seen by some as an environmental threat in its own right. In an already polarized debate over energy and climate policy, Ivanpah’s environmental controversies made it an easy target.

Engineering at scale: Maintenance, gas use, and land

Even leaving politics aside, Ivanpah faced serious engineering challenges. Its capacity factor, the share of energy produced compared to its maximum potential, lagged below 30%, far behind typical gas plants, which can exceed 60%. Meanwhile, PV solar kept getting more efficient and cheaper.

The natural gas backup system, originally intended as a limited supplement, ended up running more often than expected. That raised uncomfortable questions. How “green” was a plant that depended more than planned on fossil fuel to keep running?

Maintaining more than 300,000 heliostats in a harsh desert environment was costly and labor-intensive. Dust buildup cuts efficiency, and the extreme heat poses long-term reliability challenges for mechanical systems. The facility also demanded a huge land footprint for what, in hindsight, was modest power output compared to newer photovoltaic farms. Public perception didn’t help. From the outside, Ivanpah didn&#8…

Konten dipersingkat otomatis.

🔗 Sumber: interestingengineering.com