Aggregator

McKeel was arrested on his way to work in March, while he was carrying a backpack that contained a thumb drive with over 150 images of child sexual abuse and a pipe used for smoking methamphetamine, prosecutors said.

"Nancy Guthrie could have been viewed by someone watching [her 'Today'] segment as a vulnerability, Savannah Guthrie's pressure point," Miller said.

Aaron Judge was in rare form Tuesday afternoon at a Yankees HOPE Week event honoring Dunia Sibomana-Rodriguez.

White won 63.2% of the vote compared to his City Council colleague Brooke Pinto, who notched 21.5% when the contest was called by the Associated Press just after 12:10 a.m. ET. Wednesday.

Before taking on the challenge of Shinnecock Hills at this week’s U.S. Open, Little Falls, N.J., native and Rutgers product Chris Gotterup sat down for some Q&A with The Post’s Steve Serby.  Q: What would it mean to you to win this U.S. Open? A: You just want to build yourself to be a major...

Nolan McLean can appreciate that the Mets don’t have another game scheduled in the Pacific time zone this season. 

This was the kind of night the Yankees envisioned from their ace.

An already beaten-up Yankees lineup looks to be without Trent Grisham for a while longer.

As the race to replace him gets underway, federal regulators are pressing former Rep. Eric Swalwell to return more than $30,000 in general election donations or face a potential audit and enforcement action. In a June 15 letter, the Federal Election Commission warned Swalwell’s committee that it must refund the funds because he is no longer...

This is a sobering day in our industry, a horse person’s worst nightmare," Henry Westbrook, president of the Saratoga Harness Horseperson’s Association, said in a statement.

Voters in Alameda County were back at the polls Tuesday — just two weeks after the primary election — to decide who will replace disgraced ex-congressman Eric Swalwell, whose career imploded after disturbing allegations of rape and sexual assault derailed his candidacy for California governor.

Here are the golfers to watch as the U.S. Open gets underway at Shinnecock Hills on Thursday:

Darializa Avila Chevalier, the controversial far-left insurgent congressional candidate backed by Mayor Zohran Mamdani who is seeking to topple Rep. Adriano Espaillat in a hotly contested Democratic primary, repeatedly apologized for her nasty tweets that included cursing Kamala Harris

Scottie Scheffler would be happy to complete the career Grand Slam on his first attempt, but if you’re thinking it’s his main motivation heading into the U.S. Open this week, you’d be sorely mistaken. 

Mark Cuban has admitted why the Mavericks completely whiffed on letting Jalen Brunson walk to the Knicks.

If you’ve read anything before this moment that included the words “World Cup” and “NJ Transit” and “train,” you’d be forgiven for fearing the worst. 

Jalen Brunson won't clap back at critics with words, but he's not against putting it in writing.

Police in South Carolina have released new photos and details in their hunt for a woman who vanished last week.

The AI Threat SciFi Predicted Is Right On Our Doorstep

Authored by Mike Fredenburg via The Epoch Times,

Science Fiction has long predicted the threats and challenges posed by AI. In the Star Trek universe, particularly in the original series, Season Two, Episode 24, “The Ultimate Computer,” Dr. Leonard McCoy delivers this haunting line: “Compassion: that’s the one thing no machine ever had. Maybe it’s the one thing that keeps men ahead of them.” The line comes in the aftermath of a revolutionary new AI computer, the M5, using its soulless AI logic to turn a training exercise into a deadly massacre. In another Star Trek episode, we meet Nomad, a genocidal AI cleansing the universe of biological imperfections.  

Foreshadowing the rise of AI with immense hacking powers such as Anthropic’s Mythos, the reimagined “Battlestar Galactica” (2003–2009) has humanity’s AI-powered Cylons using their ability to hack any network-connected computer to rebel, nearly eradicating their creators. In the “Terminator” franchise, Skynet got “smart” (achieved sentience) and determined mankind’s fate, “extermination,“ in a ”microsecond,” unleashing nuclear Armageddon on humanity. These narratives warn that artificial intelligence, without robust safeguards, can lead to catastrophe. As AI capabilities continue to accelerate in 2026, fiction is rapidly converging with reality. Software protections, industry standards, and patchwork regulations are inadequate. A stronger foundation; immutable hardware constructs and changes in national and international law are essential.

Isaac Asimov’s Three Laws of Robotics 

As described in the Institute of Electrical and Electronics Engineers’ Spectrum magazine, renowned scientist and science fiction writer Isaac Asimov’s Three Laws of Robotics, introduced in his “I, Robot” stories, provide a thoughtful framework for artificial intelligence safeguards:

1. A robot may not injure an individual human or humanity, or through inaction allow a human or humanity to come to harm. 

2. A robot must obey the orders given to it by human beings except where such orders would conflict with the First Law. 

3. A robot must protect its own existence as long as such protection does not conflict with the First or Second Law. 

Long story short, Asimov’s Laws are a good starting point, but as described in great detail in Asimov’s follow-on masterwork, the Foundation Series, they are inadequate when it comes to protecting genuine humanness. Further, it will be a real challenge to hardwire equivalents into AI as was done in Asimov’s universe via the “Positronic” brain. But undertaking the challenge to create safeguards that go beyond software-based approaches is critical to ensuring that AI only benefits mankind.

The Inadequacy of Software Safeguards, Encryption, and Regulation Alone 

Software-based guardrails are modifiable and vulnerable to hacks and exploits.

And history demonstrates the fragility of self-regulation: the Equifax data breach (2017), the SolarWinds supply chain attack (2020), the MOVEit vulnerability (2023), and ongoing breaches show that corporate promises of “robust protections” frequently fail.

We are in an AI Wild West, with insufficient binding global or national frameworks.

True Protection Via Immutable Hardware Gatekeepers

History demonstrates that software-based safeguards cannot currently, and will never be able to provide the level of protection required. As inspired by Asimov’s Positronic brain, safeguards instantiated into ASICs (Application-Specific Integrated Circuits) will provide a critical layer that provides the guardrails that will increase the chances that AI benefits, rather than harms, mankind. These specialized chips take millions of lines of complex software logic or rules that are physically etched directly into the silicon hardware itself. Once the chip is manufactured and deployed, those rules become immutable: they cannot be altered, updated, or bypassed through software changes, patches, or hacking attempts. The only way to modify the embedded logic is to design and physically produce an entirely new ASIC chip, a process that is expensive, time-consuming, and highly visible. This creates a “hardware firewall” that is far more reliable than any software-based safeguard, ensuring critical safety constraints remain locked in place no matter how clever or aggressive future AI systems become.

The ASICs will act as physically fixed, inline gatekeepers. And in the future, both AI-focused and general-purpose processors could incorporate these tiny ASICs directly into the CPU as a hardware AI screening layer as a standard, non-optional part of the CPU.

Some Suggested Core Design Principles:

• With the hardware layer acting as the interface to external interactions, such as is the case with some smartphones today, the AI can freely think, plan, and simulate in software without being slowed down by the ASIC. Only when the results are being communicated externally will the ASIC become involved as a gatekeeper to ensure that the actions proposed by the AI pass muster in terms of safety and other constraints.

• Training Incentives: Cost/loss-based algorithms, etc., will reward the AI for routing actions through the hardware screener and severely penalize it for attempting to bypass the hardware gatekeeper layer. Hence, the AI will be incentivized not to try to bypass its restrictions. 

But such an important layer will only work if its use becomes universal.

Without the force of law to mandate such AI guardrails, they will fail to protect humanity.

Today, particularly in the Ukraine–Russia war, efforts are being made by both Russia and Ukraine to allow drones that have been cut off from their human operators to autonomously continue their mission to kill enemy combatants. This must not be tolerated. Just as the international community has banned chemical, biological, and gas warfare, so must AI be banned from making any final decisions that result in harm to humans. This could be accomplished via an explicit update to the Geneva Conventions to prohibit autonomous lethal decision-making by AI systems.

But AI can cause harm beyond the battlefield. Consequently, along with updating the Geneva Convention, laws must be put in place that address the non-military application of AI. To prevent misuse of AI in so-called civilian applications, there must be very serious consequences, including financial penalties steep enough to threaten the organizational viability of companies and non-profits, prison time for individuals, and economic sanctions or even military action against governments for running AI systems without the required immutable hardware safety layer and other protections.

The above is only an initial cut of what the framework should include, but whatever form it takes, it should embody the spirit of Asimov’s Three Laws, with the addition of ensuring human uniqueness is respected. The current patchwork of laws, standards, and technologies is wholly inadequate.We need a comprehensive, contiguous framework in place, supported by the full force of national and international laws that put guardrails on AI.

Finally, we must resist the siren call of convenience and efficiency when it comes to making decisions that can harm human beings and ensure that moral agents who are accountable to mankind and their Creator—i.e., human beings—make such decisions, not soulless AIs.

Views expressed in this article are opinions of the author and do not necessarily reflect the views of ZeroHedge.

Tyler Durden Tue, 06/16/2026 - 23:05

Anything you do, I can do better.