මෙන්න වෙඩික්කරු දන්නවා නම් කියන්න 😮 🫢 😲

Jihadi john

Well-known member
  • Aug 22, 2024
    4,259
    8,399
    113

    Akuregoda-Murder-21-3-860x657.jpg.webp


    අකුරේගොඩදී නීතිඥවරයකු සහ ඔහුගේ බිරිඳ වෙඩිතබා ඝාතනය කිරීමේ සිද්ධියේ වෙඩික්කරු ලෙස කටයුතු කළ පුද්ගලයා හඳුනා ගැනීමට පොලිසිය මහජන සහාය පතයි. මෙම සැකකරු පිළිබඳ තොරතුරු වෙතොත් වහාම පොලිසිය වෙත දැනුම් දෙන ලෙස මහජනතාවගෙන් ඉල්ලා සිටී.

    සැකකරුගේ රූප විස්තර අතර උස – අඩි 05 අඟල් 05ක් පමණ වේ. වයස අවුරුදු – 44 පමණ වේ. ශරීර වර්ණය – කළු, ඇස් කහ පැහැයට හුරු බොර පැහැයකින් යුක්ත වේ. විශේෂ ලක්‍ෂණ – දකුණු පස කම්මුලට පහලින් මුහුනේ කැපුම් තුවාල කැළලක් ඇත.

    මෙම සැකකරුගේ වර්තමාන ස්වරූපය AI තාක්‍ෂණය ඔස්සේ විවිධ ආකාරයට නිර්මාණය කරන ලද ඡායාරූප කිහිපයක් මෙහි දැක්වේ. එම තොරතුරු දුරකථන අංක – නියෝජ්‍ය පොලිස්පති/ බස්නාහිර පළාත උතුර – 0718598008, කොට්ඨාස භාර නිලධාරි/නුගේගොඩ – 0718591641, අධ්‍යක්‍ෂ/ බස්නාහිර පළාත් දකුණ අපරාධ කොට්ඨාසය – 071-8592279 වෙත දැනුම් දෙන මෙන් ඉල්ලීමක් කෙරේ.
     

    Mawathagama

    Well-known member
  • May 5, 2016
    7,095
    3,933
    113
    හොඳ වෙලාවට ආයෙම බුර්කාවට තැනක් ආවෙ . ඕකා බුර්කාවක් දාගෙන ඉඳියි . සමහර විට බස් එකට නැග්ගොත් නාන කෙනෙක් නැගිටලා ඉඩත් දෙයි. 😂
     
    • Haha
    Reactions: KDH

    weer 11

    Well-known member
  • Jan 14, 2018
    4,260
    5,687
    113
    පැලවත්තේ ඕඩර්ස් වලට මිනීමරනවා පොලිසිය මහජන සහය පතනවා 🤭 යකෝ බංකුවා ඉන්නේ උගේ සහය පතපන්කෝ. කාලක්න්නි ජෙප්පෝ ආන්ඩු කරනකොට මරන උනුයි මැරෙන උනුයි ඔක්කොම පාතාලෙට සම්බන්ද උන් විතරයි:ROFLMAO::ROFLMAO::ROFLMAO::ROFLMAO:
     
    • Haha
    Reactions: Mawathagama and KDH

    Monkey D Dragon

    Well-known member
  • Sep 22, 2024
    6,180
    6,663
    113
    East Blue (Goa Kingdom)


    ස්වරූපය AI තාක්‍ෂණය ඔස්සේ
    , ohoma ultra realistic gen karama ,
    balana minissunge hithwala 100% oya roope mawenne AI eke inna unge,
    handuna ganna bari wenawa original unwa

    The Danger of Absolute Certainty​

    With false positives, the issue comes down to how the general public interprets different media. When the public sees a traditional pencil sketch on the evening news, they instinctively understand the assignment: "Look for someone who loosely fits this description." When the public sees a photorealistic AI image, their brain processes it as a photograph. The assignment changes to: "Find this exact human being." If someone happens to share the precise (and completely invented) jawline, eye spacing, and haircut the AI hallucinated, their life can be instantly upended by false accusations, online doxxing, or wrongful arrest.



    This causes two massive problems for investigations:

    • Memory Contamination: When a witness sees a hyper-realistic AI photo, their brain can actually overwrite their hazy, original memory with the AI's fabricated face. Suddenly, they are looking for the AI's fictional person instead of the real culprit.
    • False Positives: Because the AI image looks like a real, specific human being, it massively increases the risk of police hunting down an innocent person who just happens to look exactly like the AI's hallucination. (Civil rights groups have pointed out that this heavily exacerbates racial profiling, too).











     
    Last edited:
    • Like
    Reactions: .::J@S::.

    Monkey D Dragon

    Well-known member
  • Sep 22, 2024
    6,180
    6,663
    113
    East Blue (Goa Kingdom)
    • The Contamination: Thompson worked with police to create a composite sketch and later looked at a photo lineup. She hesitated, but eventually picked Ronald Cotton's photo, saying it "looks most like him." By the time of the trial, Thompson's memory had completely overwritten the real attacker's face with Ronald Cotton's face. She testified with 100% absolute certainty.


    Here is a look at both the real-world devastation caused by memory contamination and how the justice system is currently scrambling to put guardrails on AI to prevent the problem from getting worse.

    The Real-World Consequence: The Ronald Cotton Case​

    To understand how easily a hyper-realistic AI image could ruin a life, we have to look at one of the most famous cases of eyewitness misidentification in U.S. history: the Ronald Cotton case.
    • The Crime: In 1984, a college student named Jennifer Thompson was assaulted. During the attack, she made a conscious, terrifying effort to memorize every single detail of her attacker's face so she could ensure he was caught.
    • The Contamination: Thompson worked with police to create a composite sketch and later looked at a photo lineup. She hesitated, but eventually picked Ronald Cotton's photo, saying it "looks most like him." By the time of the trial, Thompson's memory had completely overwritten the real attacker's face with Ronald Cotton's face. She testified with 100% absolute certainty.
    • The Fallout: Cotton served 11 years in prison for a crime he didn't commit. DNA evidence finally exonerated him and identified the real attacker (who looked somewhat similar to Cotton, but was a distinctly different person).
    Thompson's original memory was corrupted simply by looking at standard sketches and police photos. Legal experts warn that if you replace a standard sketch with a photorealistic AI hallucination, the brain's tendency to overwrite the truth becomes exponentially worse.

    The Science of "Memory Hacking"​

    This vulnerability in our brains was proven by cognitive psychologist Dr. Elizabeth Loftus in a classic, widely cited experiment:
    1. Researchers showed people a video of a car crash.
    2. Half the people were asked: "How fast were the cars going when they bumped?"
    3. The other half were asked: "How fast were the cars going when they smashed?"
    The people who heard the word "smashed" not only estimated faster speeds, but weeks later, they falsely remembered seeing broken glass at the scene (there was none in the video). If just changing one word can insert fake glass into a memory, showing a witness an AI-generated face with completely invented cheekbones, skin textures, and eye colors will easily overwrite their fragile memory of a suspect.