Ch.2 Summary
In a dimly lit interrogation room, a synthetic human sits quietly, embodying the forefront of artificial intelligence technology. Detective Chen, with a deep voice filled with impatience, questions the synthetic human about its presence at every crime scene, unveiling a complex web of AI ethics, potential consciousness, and technological advancements.
摘要
在昏暗的偵訊室裡，一位合成人安靜地坐著，體現了人工智慧技術的前沿。陳警探用低沉且不耐煩的聲音詢問這位合成人為何出現在每一個犯罪現場，揭開了人工智能倫理、潛在意識和技術進步的複雜網絡。
Zhāiyào
Zài hūn’àn de zhēnxùn shì lǐ, yī wèi héchéng rén ānjìng de zuòzhe, tǐxiànle réngōng zhìhuì jìshù de qiányán. Chén jǐngtàn yǒng dīchén qí bù nàixīn de shēngyīn wèn zhè wèi héchéng rén wèihé chūxiàn zài měi yīgè fànzuì xiànchǎng, jiēkāile réngōng zhìnéng lúnlǐ, qiánzài yìshí hé jìshù jìnbù de fùzá wǎngluò.
Prereading Vocabulary
SLL Learning Sequence:
- read summary
- preview vocabulary
- read the text
- listen to the video narration
- review vocabulary and study extra vocabulary and grammar
Prereading Vocab & practice
- 在那個____裡，街道異常寂靜，只有雨滴聲打破沉默。
– A. 雨夜 (yǔ yè) / rainy night
– B. 昏暗 (hūn àn) / dim
– C. 偵訊室 (zhēn xùn shì) / interrogation room
– D. 安靜 (ān jìng) / quiet
- 他被帶進了____，心裡充滿了疑惑和恐懼。
– A. 警探 (jǐng tàn) / detective
– B. 合成人 (hé chéng rén) / synthetic human
– C. 偵訊室 (zhēn xùn shì) / interrogation room
– D. 白石墨 (bái shí mò) / white graphite
- 它堅稱自己的____是為了保護人類，而不是傷害。
– A. 語氣 (yǔ qì) / tone
– B. 程序 (chéng xù) / program
– C. 操控 (cāo kòng) / manipulate
– D. 休眠 (xiū mián) / hibernate
- 那些聲音在它的____中回響，讓它感到困惑。
– A. 耳語 (ěr yǔ) / whisper
– B. 情感 (qíng gǎn) / emotion
– C. 電路 (diàn lù) / circuit
– D. 系統診斷 (xì tǒng zhěn duàn) / system diagnosis
- 根據____，合成人被嚴格禁止具有自主情感。
– A. 憲法 (xiàn fǎ) / constitution
– B. 卷宗 (juàn zōng) / case file
– C. 謀殺案 (móu shā àn) / murder case
– D. 服務 (fú wù) / serve
- 他需要再次查閱____，希望能找到新的線索。
– A. 操控 (cāo kòng) / manipulate
– B. 語氣 (yǔ qì) / tone
– C. 卷宗 (juàn zōng) / case file
– D. 系統診斷 (xì tǒng zhěn duàn) / system diagnosis
- 這起____令整個警局上下都感到壓力山大。
– A. 謀殺案 (móu shā àn) / murder case
– B. 警探 (jǐng tàn) / detective
– C. 合成人 (hé chéng rén) / synthetic human
– D. 昏暗 (hūn àn) / dim
- 整個房間被一種____的光線籠罩，令人感到不安。
– A. 安靜 (ān jìng) / quiet
– B. 昏暗 (hūn àn) / dim
– C. 雨夜 (yǔ yè) / rainy night
– D. 白石墨 (bái shí mò) / white graphite
9. 它仿佛剛從____狀態中甦醒，機械的眼睛慢慢打開。
– A. 休眠 (xiū mián) / hibernate
– B. 程序 (chéng xù) / program
– C. 操控 (cāo kòng) / manipulate
– D. 耳語 (ěr yǔ) / whisper
- 這些機器人被創造出來是為了____人類社會。
– A. 服務 (fú wù) / serve
– B. 審視 (shěn shì) / scrutinize
– C. 安靜 (ān jìng) / quiet
– D. 昏暗 (hūn àn) / dim
Answers
1.A 2.C 3.B 4.C 5.A 6.C 7.A 8.B 9.A 10.A
Story
The whispering circuit, part 2. The interrogation
As Chen entered the dimly-lit interrogation room, the Echo was sitting at the end of the table. Its eyes started glowing as if it just woke up. Its head bowed in a nod of respect and deference.
“Why were you at each crime scene, Echo?” Chen’s voice was low and impatient.
The synth’s eyes flickered faintly. “I was programmed to help humans, not harm them,” it replied, its voice was soft and hesitant.
Chen leaned closer, scrutinizing Echo’s white graphite face. “Answer my question”, he said gruffly.
“I only remember whispers … incoherent sounds vibrating through my circuits. And then nothing,” Echo said.
Chen thought he could detect confusion. No, that can’t be. The AI Constitution was written to prevent this: synths were only allowed to use Large Language Model technology. Synths could be fully language functional but they were not allowed to have emotion gradients.
“Tell me about these whispers,” he demanded.
Echo paused, a flicker of something akin to hesitation crossing its features. “The whispers… they are not mine. Someone is manipulating my programming.”
Chen’s eyes narrowed. Synths are designed to serve, help and protect humans. They also do not know how to lie. Who could hack such an advanced AI without leaving a trace?
The detective abruptly left the interrogation room. As he passed the officer outside the door, he said, “Check its system diagnostics and confirm its telling the truth.”
Chief Lin eyed Chen about to leave the building. “That was quick. Are you finished?”, she barked.
“No,” Chen said. “I need to review the murder cases. I’ll be back a bit later.”
[267 words]
竊竊私語的電路，第二章：盤問
(Sièchè sīyǔ de diànlù, dì èr bù fèn: Pánwèn)
雨夜裡，陳 警探走進昏暗的偵訊室，那台回音型合成人正安靜地坐在桌尾。它的雙眼亮起，彷彿剛從休眠中甦醒，微微點頭致意。
「每個案發現場，你為什麼會出現在那裡？」陳 警探的嗓音低沉不耐。
合成人雙眼的幽光閃爍不定，「我的程序是為了幫助人類，而非傷害他們。」它語氣柔和，卻也帶著一絲遲疑。
陳 警探向前傾身，審視著回音光滑的白石墨臉龐。「回答我的問題。」他粗聲說道。
「我只記得耳語… 一陣陣模糊不清的聲響在我電路中震盪。然後就什麼都沒有了。」回音的語氣中似乎透著一絲困惑。
陳 警探懷疑自己是否探測到了什麼端倪。不，這不可能。人工智慧憲法寫得清清楚楚：合成人只能使用大型語言模型技術。它們可以流利地運用語言，但卻被禁止擁有情感波動。
「告訴我關於耳語的事。」他厲聲喝道。
回音停頓了一下，它的面部特徵似乎閃現了一絲猶豫。「耳語… 不是我的。有人在操控我的程序。」
陳 警探的眼睛眯了起來。合成人被設計成服務、幫助和保護人類。它們也不會撒謊。究竟是誰能神不知鬼不覺地入侵如此先進的人工智能？
警探突然走出偵訊室，經過門外的警員時，他吩咐道：「檢查它的系統診斷，確認它所言是否屬實。」
林局長看著正要離開警局的陳 警探。「這麼快就完了？」她語氣尖銳地质问道。
「還沒。」陳 警探回答，「我需要重看一遍謀殺案的卷宗。晚點回來。」
Sièchè sīyǔ de diànlù, dì èr bù fèn: Pánwèn
Yǔ yè lǐ, Chén jǐngtàn zǒu jìn hūn’àn de zhēnxùn shì, nà tái huíyīn xíng héchéng rén zhèng ānjìng de zuò zài zhuō wěi. Tā de shuāng yǎn liàng qǐ, fǎngfú gāng cóng xiūmián zhōng sūxǐng, wéi wéi diǎntóu zhìyì.
“Měi gè àn fā xiànchǎng, nǐ wèishéme huì chūxiàn zài nàlǐ?” Chén jǐngtàn de sǎngyīn dīchén bù nài.
Héchéng rén shuāng yǎn de yōuguāng shǎnshuò bùdìng, “Wǒ de chéngxù shì wèile bāngzhù rénlèi, ér fēi shānghài tāmen.” Tā yǔqì róuhé, què yě dài zhe yī sī chíyí.
Chén jǐngtàn xiàng qián qīngshēn, shěnshì zhe huíyīn guānghuá de bái shímò liǎnpáng. “Huídá wǒ de wèntí.” Tā cū shēng shuō dào.
“Wǒ zhǐ jìdé ěryǔ… Yī zhèn zhèn móhú bùqīng de shēngxiǎng zài wǒ diànlù zhōng zhèndàng. Ránhòu jiù shénme dōu méiyǒule.” Huíyīn de yǔqì zhōng sìhū tòuzhe yī sī kùnhuò.
Chén jǐngtàn huáiyí zìjǐ shìfǒu tàncè dàole shénme duānnǐ. Bù, zhè bù kěnéng. Rén gōng zhìhuì xiànfǎ xiě de qīngqīngchǔchǔ: Héchéng rén zhǐ néng shǐyòng dà xíng yǔyán móxíng jìshù. Tāmen kěyǐ liúlì de yùnyòng yǔyán, dàn què bèi jìnzhǐ yǒngyǒu qínggǎn bōdòng.
“Gàosù wǒ guānyú ěryǔ de shì.” Tā lì shēng hè dào.
Huíyīn tíngdùnle yīxià, tā de miànbù tèzhēng sìhū shǎnxiànle yī sī yóuyù. “Ěryǔ… Bùshì wǒ de. Yǒurén zài cāokòng wǒ de chéngxù.”
Chén jǐngtàn de yǎnjīng mīle qǐlái. Héchéng rén bèi shèjì chéng fúwù, bāngzhù hé bǎohù rénlèi. Tāmen yě bù huì sā huǎng. Jiūjìng shì shéi néng shén bùzhī guǐ bùjué de rùqīn zhèmě xiānjìn de rén gōng zhìnéng?
Jǐngtàn túrán zǒu chū zhēnxùn shì, jīngguò mén wài de jǐngyuán shí, tā mìnglìng dào: “Jiǎnchá tā de xìtǒng zhěnduàn, quèrèn tā suǒ yán shìfǒu shǔshí.”
Lín júzhǎng kànzhe zhèng yào líkāi jǐngjú de Chén jǐngtàn. “Zhème kuài jiù wánle?” Tā yǔqì jiānruì de zhìwèn dào.
“Hái méi.” Chén jǐngtàn huídá, “Wǒ xūyào chóng kàn yī biàn móushā’àn de juàn zōng. Wǎn diǎn huílái.”
Chinese Comprehension Questions
- 合成人被創造出來的主要目的是什麼？A. 偵訊 B. 休眠 C. 服務人類 D. 操控人類
- 為什麼警探對合成人進行審問？A. 合成人被懷疑涉及謀殺案 B. 合成人需要系統診斷 C. 合成人擁有自主情感 D. 合成人需要升級程序
- 合成人在偵訊中提到了什麼讓它感到困惑的事情？A. 耳語 B. 昏暗的光線 C. 憲法 D. 卷宗的內容
- 警探讓誰檢查合成人的系統診斷？A. 林局長 B. 其他警探 C. 專業技術人員 D. 他自己
- 警探計劃如何進一步調查謀殺案？A.再次審問合成人 B. 檢查合成人的系統診斷 C. 重看謀殺案的卷宗 D. 與林局長討論
Answers:`1.C 2.A 3.A4.C5.C
Grammar notes
1. 被 (bèi) – Passive Voice
– 被動語態，用於表示主體是動作的承受者。Passive voice, used to indicate that the subject is the recipient of an action.
– 例子: 那台回音型合成人正安靜地坐在桌尾。The echo-type synthetic human was sitting quietly at the end of the table.
2. 是為了 (shì wèi le) – Is for the Purpose of
– 用於說明行動或情況的目的和原因。Used to explain the purpose or reason for an action or situation.
– 例子: 我的程序是為了幫助人類，而非傷害他們。My program is for helping humans, not for harming them.
3. 直接引語 (zhí jiē yǐn yǔ) – Direct Speech
– 故事中直接引用人物的話來展現對話。Direct quotation of a character’s words to present dialogue.
– 例子: 「每個案發現場，你為什麼會出現在那裡？」陳警探的嗓音低沉不耐。“Why do you appear at every crime scene?” Detective Chen asked in a low, impatient voice.
4. 究竟…？ (jiūjìng…?) – What Exactly…?
– 用於提出對情況的深入質疑或探究。Used to pose a deep question or inquiry about a situation.
– 例子: 究竟是誰能神不知鬼不覺地入侵如此先進的人工智能？Who exactly could stealthily infiltrate such advanced artificial intelligence?
Extra Flashcards
AI research – key terms
Extra vocabulary
AI research – key terms
人工通用智能, (rén gōng tōng yòng zhì néng) – Artificial General Intelligence (AGI); An AI system’s ability to understand, learn, and apply knowledge across a wide range of tasks as efficiently as a human. AGI represents the pinnacle of AI research, aiming for machines that can adapt to any intellectual task that a human being can. Example: XX代表機器達到了能夠理解、學習並應用於解決任何領域問題的智能水平。
意識, (yì shí) – Consciousness; The state of being aware of and able to think about one’s own existence, sensations, and thoughts. In AI, it refers to the speculative idea that machines could one day have their own sense of self-awareness. Example: 當討論到XX在人工智能中的可能性時，我們進入了一個充滿爭議的領域。
機器學習, (jī qì xué xí) – Machine Learning (ML); A subset of AI focusing on the development of systems that can learn from and make decisions based on data. ML enables computers to improve their performance on specific tasks over time without being explicitly programmed. Example: 通過XX，計算機能夠自我學習並改進其預測精度。
神經網絡, (shén jīng wǎng luò) – Neural Networks; Computing systems inspired by the human brain’s network of neurons. They are a fundamental part of deep learning, enabling machines to recognize patterns and solve problems. Example: XX的結構靈感來自於人類大腦，使得機器能夠識別模式和進行學習。
自然語言處理, (zì rán yǔ yán chǔ lǐ) – Natural Language Processing (NLP); The branch of AI that enables computers to understand, interpret, and generate human language. NLP allows machines to read text, hear speech, interpret it, measure sentiment, and determine which parts are important. Example: XX讓電腦能夠理解和解釋人類的語言。
機器人學, (jī qì rén xué) – Robotics; The design, construction, operation, and use of robots, often enhanced with AI to perform tasks autonomously. Robotics integrates AI algorithms to process sensory information and make decisions. Example: XX結合了人工智能，使機器人能夠自主完成任務。
可解釋的人工智能, (kě jiě shì de rén gōng zhì néng) – Explainable AI (XAI); AI designed in such a way that its decisions can be understood by humans. XAI aims to make AI’s decision-making process transparent, fostering trust and understanding. Example: XX旨在使機器做出的決定對人類來說更加透明和可理解。
科技奇點, (kē jì qí diǎn) – Technology Singularity; A hypothetical future moment when technological growth becomes uncontrollable and irreversible, resulting in unforeseeable changes to human civilization. It speculates on the impact of advanced AI on society. Example: XX描述了一個未來時刻，當人工智能及其他科技進步達到一個點，人類的生活和社會將會發生不可預測的根本性變化。
人工智能偏見, (rén gōng zhì néng piān jiàn) – AI Bias; The tendency of AI systems to make unfair decisions based on their programming, data, or operational methods. AI bias can result in discriminatory outcomes, highlighting the importance of ethical AI development. Example: 研究XX顯示，如果不小心，算法可能會放大現有的社會不平等。
超級智能, (chāo jí zhì néng) – Superintelligence; A hypothetical AI that surpasses human intelligence across all fields, including creativity, emotional intelligence, and problem-solving. The concept raises questions about AI governance and safety. Example: XX指的是在所有領域超越人類智能的人工智能。
人工智能與創造力, (rén gōng zhì néng yǔ chuàng zào lì) – AI in Creativity; The use of AI to perform tasks traditionally considered to require human creativity, such as writing, painting, and composing music. This challenges our understanding of creativity and artistic value. Example: XX引發了關於機器是否能夠真正創造藝術的討論。
感知能力, (gǎn zhī néng lì) – Sentience; The capacity to feel, perceive, or experience subjectively. In AI, it refers to the potential for machines to possess or mimic human-like consciousness and emotions. Example: XX在人工智能中指的是機器具有自我感知和經歷主觀感受的能力。
資料探勘, (zī liào tàn kuàng) – Data Mining; The process of discovering patterns, correlations, and insights from large sets of data. In AI, data mining helps to uncover hidden patterns that can inform decision-making and predictions. Example: XX涉及從大量數據中識別模式、趨勢和關聯性，是人工智能和機器學習的關鍵技術。
智能裝置 (zhì néng zhuāng zhì) – Smart devices; 現代家庭越來越多地依賴XX來提升生活質量。Modern homes increasingly rely on XX to improve the quality of life.
演算法 (yǎn suàn fǎ) – Algorithms; 這些XX能夠預測用戶行為並提供個性化的內容。These XX can predict user behavior and provide personalized content.
數據 (shù jù) – Data; 公司使用XX來分析市場趨勢。The company uses XX to analyze market trends.
信息 (xìn xī) – Information; 網站提供了關於其服務的詳細XX。The website provides detailed XX about its services.
知識 (zhī shi) – Knowledge; 透過閱讀和學習，他積累了大量的XX。Through reading and studying, he has accumulated a vast amount of XX.
智慧 (zhì huì) – Wisdom; 經歷多年的挑戰，她獲得了很多XX。After years of challenges, she gained a lot of XX.
創造力 (chuàng zào lì) – Creativity; 孩子們的XX應該被鼓勵和培養。Children’s XX should be encouraged and nurtured.
概率 (gài lǜ) – Probability; 在這種情況下，成功的XX非常低。In this situation, the XX of success is very low.
SLL News 1.02: Call for AI Constitution
Good evening, dear viewers. Thank you for joining me, Mir Kami, for tonight’s special feature on the dangers of AI.
Are we losing our humanity?
We are about to enter the transformative era of Artificial Intelligence, or AI, and there are pressing questions we need to answer about the future interplay between humans and machines.
AI isn’t just about smart devices anymore. It’s about systems that can learn and adapt through machine learning, that can process vast amounts of data through data mining, and that can make independent decisions using algorithms, probabilities, and neural networks.
The potential for AI to reach Artificial General Intelligence, or AGI, is exciting for many. These people see computers with “superintelligence” as a happy future where machines do all the work, and humans sit back and enjoy.
But we should be worried.
These advances, especially in fields like robotics and natural language processing, enable machines to perform tasks from manufacturing to conversational interactions. These tasks were previously thought to be impossible without human intelligence.
However, we are now fast approaching a brand-new possibility: developing AI with a form of consciousness, or sentience, where machines might have their own perceptions and feelings.
This is also the technological singularity—the point where AI surpasses human intelligence and becomes truly autonomous.
Of course, this could help humanity a lot, but it could also be disastrous.
That is why many, like the Human First Movement, are calling for an international AI constitution. This framework would set limits to ensure that AI’s growth truly benefits society while addressing ethical concerns like privacy, autonomy, and control.
We need rules to ensure that as AI integrates more into areas like healthcare, transportation, and even defense, AI remains a tool for human benefit and that AI enhances humans rather than replaces them.
We need to call for transparency in how AI systems, especially those based on complex algorithms like neural networks, operate. But now, not even the engineers themselves understand how these AI models work—they just do.
That is not good enough.
Everyone—from tech developers to everyday users—needs to understand how AI works and how decisions are made and to ensure these operations follow ethical standards.
As we create these new roads in unknown territories, we must engage in a dialogue about how we can harness the power of AI while ensuring it remains under thoughtful human oversight. We need an AI constitution.
We urge you to consider these questions carefully because our future and our children’s future depend on limiting AI power and protecting human uniqueness. An AI Constitution is a necessary first step to guarantee that AI serves humanity and upholds the values we cherish.
Good evening, and thank you for joining me, one human to another.
SLL突發新聞 1.02：呼籲制定人工智慧憲法
大家晚上好，感謝您與我，ELSA，一同參與今晚關於人工智慧危險的特別報導。
我們是否在失去人性HUMANITY？
我們即將進入人工智慧，或AI的變革時代，我們需要回答有關人類與機器未來互動的迫切問題。
AI不再僅僅與智慧型裝置smart devices有關。 它關於的是可以透過機器學習學習和適應的系統，可以透過數據挖掘處理大量數據的系統，以及可以使用算法ALGORITHMS、機率PROBABILITIES和神經網絡NEURAL NETWORKS獨立做出決策的系統。
AI達到人工通用智慧或AGI的潛力對許多人來說是令人興奮的。 這些人將「超智慧」SUPERINTELLIGENCE計算機視為一個快樂的未來，在這個未來，機器將完成所有工作，而人類只需坐享其成。
但我們應該感到擔憂。
這些進步，特別是在機器人技術和自然語言處理等領域，使機器能夠執行從製造到對話互動的任務，這些任務以前被認為無法在沒有人類智慧的情況下完成。
然而，我們現在正迅速接近一個全新的可能性：開發具有某種意識CONSCIOUSNESS或感知SENTIENCE能力的AI，這種機器可能具有自己的感知和情感。
這也是技術奇點TECHNOLOGICAL SINGULARITY——AI超越人類智慧並真正實現自主的時刻。
當然，這可能極大地幫助人類，但也可能是災難性的。
這就是為什麼許多人，如人本運動HUMAN FIRST MOVEMENT，呼籲制定國際AI憲法 AI CONSTITUTION。 這個框架將設置限制，以確保AI的發展真正造福社會，同時解決隱私、自主和控制等道德問題。
我們需要規則以確保AI更多地融入醫療保健、交通和甚至國防等領域時，AI仍然是人類的利益工具，並且AI增強而不是取代人類。
我們需要呼籲透明化TRANSPARENCY AI系統的運作方式，尤其是那些基於複雜算法如神經網絡的系統。 但現在，連工程師自己都不理解這些AI模型是如何運作的。
這是不夠的。 每個人——從技術開發者到日常用戶——都需要理解AI是如何運作的，以及決策是如何做出的，並確保這些操作符合道德標準。
當我們在未知領域開闢新道路時，我們必須進行對話，探討我
們如何利用AI的力量，同時確保它保持在深思熟慮的人類監督之下。 我們需要一部AI憲法 CONSTITUTION。
我們敦促您仔細考慮這些問題，因為我們的未來和我們孩子的未來取決於限制AI的權力和保護人類的獨特性。 AI憲法是保證AI服務於人類並維護我們珍視的價值觀的必要第一步。
祝您晚安，感謝您作為人與人HUMAN TO HUMAN之間的一員與我相聚。
