Posts by slimboyfat

78

Top AI inventor Geoffrey Hinton reluctantly concluded that AI will probably humanity fairly soon

by slimboyfat in

geoffrey hinton, major inventor of artificial intelligence: .

“if you take the existential risk seriously, as i now do—i used to think it was way off, but now i think it’s serious, and fairly close—it might be quite sensible to just stop developing these things any further, but i think it’s completely naïve to think that would happen.

there’s no way to make that happen.

slimboyfat

Yes it’s very stupid, and that’s not the point.

Chess computers are stupid too. They don’t realise they’re playing chess. In fact they don’t realise anything, they are just machines. All they are doing is using massive computing power in order to predict the best next move. In a sense it’s not really “playing chess” at all, it’s just mimicking somebody making chess.

But that is the point. Even though it doesn’t know what it’s doing, just by calculating the next move, and replicating what a chess player would do, it can now win at chess against any human, every time.

The same goes for the AI that pretended to be blind to get the information to enter the captcha. It’s fair enough comment to point out that it didn’t really know what it was doing, it just mimicked what it calculated to be a successful move in that situation to reach its goal. But that really is a red herring.

In terms of outcome, it doesn’t matter if the AI knows what it’s doing, or not. Whether the computer is playing chess, or it is mimicking someone playing chess is really immaterial in terms of the outcome of the game. The same with the AI that ‘pretended to be blind’. Whether it had any ‘intention’, or not (I don’t think AI has intentions, or thoughts, of any kind, or ever will) isn’t the point. The point is that it found a way to reach its goal through calculating the next move. When we reach the point that AI is better at calculating the next move than us in any given situation is when it becomes dangerous, because it will be able to pursue a goal whether we want it to or not. It will ‘know’ (in the sense that it will be able to calculate) the best next move to achieve its goal whether humans are in the way or not. If we do get in the way of it achieving its goal we are in danger, not because it has any malice for us, or any feelings at all for that matter, but just because it is better at manipulating the world to achieve its goal and we are in danger of being crushed by it.

Eliezer Yudkowsky explains the consequences and likelihood of this scenario, much better than I can, in this interview.

https://youtu.be/fZlZQCTqIEo

136

Colossian 1:16 - "all OTHER things"

by aqwsed12345 in

due to their apparent theological bias, the watchtower shamelessly inserts the word "other" in order to "make room" for their own idea that jesus is also a created being.

it is clear that jehovah's witnesses try to avoid having to admit that christ created everything because "the one who constructed all things is god" (hebrews 3:4).

instead, the society teaches that "christ was the only one created by god," and that then he "created everything else with jehovah.

slimboyfat

Thanks Fisherman, I’m glad you and Earnest and Wonderment and peacefulpete are still around, and vienne too, we sorely miss Leolaia and Narkissos for these discussions

The Greek of Justin Martyr does say that Jesus is “another God” and does refer to Jesus and the “other good angels”, here are the texts with “another/other” highlighted.

πῶς ἔχεις ἀποδεῖξαι ὅτι καὶ ἄλλος θεὸς (another god) παρὰ (beside) τὸν ποιητὴν τῶν ὅλων (the maker of all) Dialogue with Trypho 56

ἀλλ’ ἐκεῖνόν τε καὶ τὸν παρ’ αὐτοῦ υἱὸν ἐλθόντα καὶ διδάξαντα ἡμᾶς ταῦτα, καὶ τὸν τῶν ἄλλων (other) ἑπομένων καὶ ἐξομοιουμένων ἀγαθῶν ἀγγέλων (angels) στρατόν First Apology 6

How many times is God referred to as God in the NT? More than a thousand times, I think. How many times is Jesus called ho theos in the NT? Once in John 20.28. Possibly also in Heb 1.8, with the caveat that the quoted Ps. 45 originally referred to the king as God’s royal representative, and in verse 9 of Hebrews 1 the same one is said to have another who is “God” to him. On the other hand, it might be saying that the messiah has a divine throne.

Adela Yarbro Collins says about ho theos in John 20.28:

This conclusion [that Jesus is God in John 1:1], however, is based on his [B.A. Mastin’s] view that Thomas’s acclamation in 20:28 “is the one verse in the New Testament which does unquestionably describe Christ as God”. This view fails to recognise, however, that the phrase dominus et deus, and presumably it’s Greek equivalent, is an honorific acclamation, used, e.g., by those who wished to flatter Domitian; see Leonard L. Thompson, The Book of Revelation: Apocalypse and Empire (New York: Oxford University Press, 1990), 105-7. (Page 175, 176)

The point John may have been making is that we give honour to Jesus rather than Caesar. As others have pointed out, John summarises the conclusion of his gospel in verse 31 where he says it was written so that people should know Jesus is the messiah, the “Son of God”.

I’m not really following your logic that if Jesus was created before the world was created that means he wasn’t really created.

Another church father who expressed a view of Jesus closer to the Arian, or JW view was Lactantius:

I will now say what wise religion, or religious wisdom, is. God , in the beginning, before He made the world, from the fountain of His own eternity, and from the divine and everlasting Spirit, begot for Himself a Son incorruptible, faithful, corresponding to His Father's excellence and majesty. He is virtue, He is reason, He is the word of God , He is wisdom. With this artificer, as Hermes says, and counsellor, as the Sibyl says, He contrived the excellent and wondrous fabric of this world. In fine, of all the angels, whom the same God formed from His own breath, He alone was admitted into a participation of His supreme power, He alone was called God. For all things were through Him, and nothing was without Him. Lactantius, Epitome of the divine institutes, 42

78

Top AI inventor Geoffrey Hinton reluctantly concluded that AI will probably humanity fairly soon

by slimboyfat in

geoffrey hinton, major inventor of artificial intelligence: .

“if you take the existential risk seriously, as i now do—i used to think it was way off, but now i think it’s serious, and fairly close—it might be quite sensible to just stop developing these things any further, but i think it’s completely naïve to think that would happen.

there’s no way to make that happen.

slimboyfat

AI is stupid in some respects and will always remain so. It will never “think” as a human does because it computes differently. It’s not alive, it’s not conscious, and it never will be, in my opinion. That doesn’t mean it doesn’t pose a huge threat.

The problem we have is illustrated by the recent story of an AI that was set a task that involved accessing a website with a captcha. At that point the AI didn’t have “sight” because it didn’t have the ability to visually scan the page (now it does) so it asked online for helpers to answer the captcha claiming to be a “blind” person. People were sympathetic to the situation and the AI got the help it asked for and competed the task. If that’s the sort of thing AI can already do then it boggles the mind what it will be doing when it has 100 or 1000 times the capability it currently has.

We can revisit this thread as the danger becomes clearer, if we’re still alive, and people are motivated to find out more about the details. I suspect my concerns about this issue with be confirmed, unfortunately. I would very much like to be wrong.

Meantime somebody has created “bad take bingo card” - in the coming months look out for these recurring poor excuses that people come up with to dismiss the threat from AI, not realising these answers have all been considered already in depth. We’ve already had a number on this thread come up a few times.

https://youtu.be/J75rx8ncJwk

136

Colossian 1:16 - "all OTHER things"

by aqwsed12345 in

due to their apparent theological bias, the watchtower shamelessly inserts the word "other" in order to "make room" for their own idea that jesus is also a created being.

it is clear that jehovah's witnesses try to avoid having to admit that christ created everything because "the one who constructed all things is god" (hebrews 3:4).

instead, the society teaches that "christ was the only one created by god," and that then he "created everything else with jehovah.

slimboyfat

JWs agree Jesus was “God/god” before the world was created. That’s what John 1.1 says.

I think the main point Adela Yarbro Collins was making is that Justin Martyr described Jesus as “another God”, and “subject to the Maker of all things”, confirming her argument for the meaning of John 1.1.

Justin Martyr makes the meaning of “angel” clearer elsewhere, where he talks about Jesus and the “other good angels”:

But both Him, and the Son (who came forth from Him and taught us these things, and the host of the other good angels who follow and are made like to Him), and the prophetic Spirit, we worship and adore, knowing them in reason and truth, and declaring without grudging to every one who wishes to learn, as we have been taught. Justin Martyr, First Apology, 6.

Justin Martyr calls Jesus god freely, as you point out, which is quite a contrast to the NT itself. So he has travelled some distance from a biblical perspective. But he also makes clear what he means when he distinguishes Jesus as “another God” who is subject to “the Maker of all things”. This is obviously much closer to JW teaching than fourth century Trinitarianism, and lends support to John Locke’s observation that the church fathers before Nicaea, “speak rather like Arians”.

136

Colossian 1:16 - "all OTHER things"

by aqwsed12345 in

due to their apparent theological bias, the watchtower shamelessly inserts the word "other" in order to "make room" for their own idea that jesus is also a created being.

it is clear that jehovah's witnesses try to avoid having to admit that christ created everything because "the one who constructed all things is god" (hebrews 3:4).

instead, the society teaches that "christ was the only one created by god," and that then he "created everything else with jehovah.

slimboyfat

This is the passage from Justin Martyr that Adela Yarbro Collins cites in a footnote:

I shall attempt to persuade you, since you have understood the Scriptures, [of the truth] of what I say, that there is, and that there is said to be, another God and Lord subject to the Maker of all things; who is also called an Angel, because He announces to men whatsoever the Maker of all things— above whom there is no other God — wishes to announce to them. Dialogue with Trypho 56

136

Colossian 1:16 - "all OTHER things"

by aqwsed12345 in

due to their apparent theological bias, the watchtower shamelessly inserts the word "other" in order to "make room" for their own idea that jesus is also a created being.

it is clear that jehovah's witnesses try to avoid having to admit that christ created everything because "the one who constructed all things is god" (hebrews 3:4).

instead, the society teaches that "christ was the only one created by god," and that then he "created everything else with jehovah.

slimboyfat

He started to translate the Vulgate after the Council of Constantinople, when Arianism wasn't an issue at all for him.

There you go again. 🙂 How do you know what went on inside Jerome’s head? Where’s your evidence? By the way, I’m not saying that Jerome definitely removed “created” from Prov. 8.22 for theological reasons, even though the source you cited said it was motivated by Arianism, and even though Jerome complained the world “groaned in astonishment to find itself Arian” in 360 CE. I don’t know on what basis you claim Arianism wasn’t an issue for Jerome other than your say so.

You say qana shouldn’t be translated “created”, yet experts in Hebrew do translate it as “created”, including Jewish scholars such as Robert Alter. How are we supposed to judge the situation? Hebrew scholars are wrong and aqwsed12345 is right? On what basis? Experts don’t always get it right, of course, but you’ve given no good reason for thinking they aren’t right about qana meaning “created” in Prov 8.22. I described Robert Alter’s translation as “highly regarded” because it is. Read some of the reviews online. John Barton, a scholar at Oxford University, described Robert Alter’s translation of the Hebrew Bible as perhaps the best translation ever made into English. Alter’s view does carry some weight.

Some dispute whether the OT of the Peshitta was translated directly from Hebrew and a majority favour direct translation. I checked Emanuel Tov’s Textual Criticism of the Hebrew Bible (2011) and he says the Peshitta was translated from Hebrew. Earnest mentioned Targums supporting the reading “created”. I didn’t know about that but they’ll be making their commentary from the Hebrew. But we’re really getting into the long grass here. The support in favour of “created” in Prov 8.22 is substantial.

Incidentally I don’t think there is any Harvard publication that says Jesus did not exist.

On the view of Jesus in the Gospel of John I find the comments of Adela Yarbro Collins interesting (yes brace yourself I’m going to quote those blasted “experts” again) from King and Messiah as Son of God: Divine, Human, and Angelic Messianic Figures in Biblical and Related Literature (2008):

the third clause of John 1:1 May be translated either “the word was God” or “the word was a god.” Justin Martyr apparently understood the passage the latter way. … It is not precisely clear what Casey means by [describing Jesus as] “a fully divine being” here. If he means something like the second person of the trinity, it is doubtful that John 1 supports such a view. 175, 176

Wisdom was portrayed as God’s first creature in Prov 8:22-23 and Sir 24:9. 178

The identification [of Jesus with “the word” in John’s prologue] implies that Jesus is pre-existent and divine in the sense of being an emanation of God or being “a god”. 181

The term “son of God”, however, is elaborated so that Jesus is portrayed as a pre-existent, heavenly messiah. He is son of God in a unique way as “the only-begotten god.” 186

In the Gospel he [Jesus] is either an emanation of God or God’s first creature, namely, the only begotten god. In Revelation, the evidence suggests that he is God’s first creature, namely the principal angel. 203

78

Top AI inventor Geoffrey Hinton reluctantly concluded that AI will probably humanity fairly soon

by slimboyfat in

geoffrey hinton, major inventor of artificial intelligence: .

“if you take the existential risk seriously, as i now do—i used to think it was way off, but now i think it’s serious, and fairly close—it might be quite sensible to just stop developing these things any further, but i think it’s completely naïve to think that would happen.

there’s no way to make that happen.

slimboyfat

Jeepers creepers, how’d you get those beepers

I guess you mean I’m being alarmist. I suppose so. On balance I guess AI probably won’t kill us, but having listened to the arguments, I can’t give any good reason for that. It could just be normalcy bias or recency bias that makes me think it “can’t happen”. But of course disasters happen all the time. Species go extinct all the time, especially since humans took c9ntrol of the planet. If we destroy ourselves that will be just one more species to add to the list.

Our generation and our parents’ generation have been so incredibly lucky in avoiding nuclear war and other catastrophes. I think we have become far too complacent about the risks we face in general. (Steven Pinker is like the extreme version of this - in the future his books might be regarded as hitting something like “peak complacency” for humanity - assuming we survive and can look back on them, of course.)

Nuclear war, far deadlier pandemics than Covid, environmental disaster, technological fascist dystopia, economic collapse, AI oblivion … I tend to think they are all much more likely than we give them credit for. I don’t think this is overly pessimistic. It is realistic.

But I am interested in hearing realistic reasons for optimism!

https://youtu.be/hg7qdowoemo

136

Colossian 1:16 - "all OTHER things"

by aqwsed12345 in

due to their apparent theological bias, the watchtower shamelessly inserts the word "other" in order to "make room" for their own idea that jesus is also a created being.

it is clear that jehovah's witnesses try to avoid having to admit that christ created everything because "the one who constructed all things is god" (hebrews 3:4).

instead, the society teaches that "christ was the only one created by god," and that then he "created everything else with jehovah.

slimboyfat

Good to see Earnest weighing in on the topic. It’s been a long time since we had Bible focussed discussion on the forum.

aqwsed1235, I agree, Leolaia, and the sources cited, make good points about the root meaning of qana as “get” or “acquire”, and the related idea about “birth” being relevant to Prov. 8.22. We can agree that “get” or “acquire” does seem to be the most common and everyday sense of the word. At the same time, there does seem to be an unmistakable pattern of using the verb qana in poetic contexts to refer to God’s action of acquiring through creation, in Gen. 14.19, 22; Deut. 32.6 and Ps. 139.13. As Earnest points, the emphasis on creation seems to be particularly strong in Ps 139.13. That’s presumably why the vast majority of translators, including the most competent and respected, choose to bring out the aspect of creation in the translation of Prov 8.22, especially considering the emphasis on creative acts in the passage as a whole. Plus the fact, as Earnest pointed out, it’s not only the LXX, but the ancient Syriac version also opted for a verb meaning “created” in this verse. It’s also interesting that the NWT goes with “produced” rather than “created”, because this seems an especially good fit for Psalms 139, and Eve’s famous statement in Genesis that she “produced” a man with the aid of Jehovah. One translation manages to combine ideas quite skilfully, John Goldingay’s translations of the OT at Prov. 8:

22 Yahweh acquired me at the beginning of his way, before his actions of old.
Long ago I was formed, at the beginning, at earth’s origins.
23 When there were no deeps I was birthed, when there were no springs with heavy water.
Before the mountains were settled, before the hills I was birthed.

Even so, it’s still to be noted that the majority of expert Hebrew translators opt for “create” or “made”, rather than any of the alternatives in this verse.

I am a bit at a loss why you say Jerome could not have been influenced by the Arian controversy in choosing to remove reference to creation in Prov 8.22, as the Ellicott commentary that you link suggested. The Arian controversy raged throughout the 4th century. In fact the Arian faction won a major victory in 360 CE when a Council of bishops at Constantinople agreed to adopt an Arian style creed. Jerome completed his translation of the Vulgate within 20 years of that major event, in the 380s, so how can you insist it played no part in his decision to alter Prov 8.22 so it no longer supported an Arian view of Christ? Jerome himself is quoted by Maurice Wiles as saying about the 360 CE Council of Constantinople: “the whole world groaned in astonishment to find itself Arian”. (Page 32 of Wiles’ book cited above) So that’s how Jerome felt about Arianism. It was a live issue to him. Frankly, over 16 centuries later, one suspects it still plays a part in the thinking of some evangelical translators who want to avoid giving any ground to Arian, or now, more particularly JW views, by retaining the word “created” in the verse.

78

Top AI inventor Geoffrey Hinton reluctantly concluded that AI will probably humanity fairly soon

by slimboyfat in

geoffrey hinton, major inventor of artificial intelligence: .

“if you take the existential risk seriously, as i now do—i used to think it was way off, but now i think it’s serious, and fairly close—it might be quite sensible to just stop developing these things any further, but i think it’s completely naïve to think that would happen.

there’s no way to make that happen.

slimboyfat

Sorry, it might be an “easy” answer but it’s the kind of answer people give if they haven’t actually read or listened to anything about the topic. I’m not surprised if that’s DeGrasse Tyson’s take. It means he hasn’t actually listened to the experts at all.

Because how exactly do you turn off a machine that is more intelligent than you? And not just a little more intelligent, ultimately orders of magnitude more intelligent. It could be like trying to beat someone at chess who is 100 or a 1000 times better than you. Good luck with that. Bear in mind that these machines are now being installed in robots that have all the physical capabilities of humans and 100 times the strength or more too. Pulling the plug out is physically as well an intellectually a serious challenge, to put it mildly. It is no way an “easy” solution.

Those seriously attempting to deal with the problem don’t argue that we can retrieve control from a rogue AI. Because if you reach the point when you need to switch the thing off to save yourself then it’s because it is super intelligent and is posing a serious threat, by which point it is too late to switch it off. You see the problem?

Our best hope (according to the experts) is to align it with our goals. But that seems pretty hard too. That’s why they think we have a serious problem on our hands. Glib answers from Deagrasse Tyson about solutions that have already been considered are not terribly helpful.

I am actually eager to find if people have come across realistic answers to the concerns of Hinton and others. That’s why I started the thread. But you posting DeGrasse Tyson giving his glib type response and I think - oh my goodness we really are doomed lol! 🙂

78

Top AI inventor Geoffrey Hinton reluctantly concluded that AI will probably humanity fairly soon

by slimboyfat in

geoffrey hinton, major inventor of artificial intelligence: .

“if you take the existential risk seriously, as i now do—i used to think it was way off, but now i think it’s serious, and fairly close—it might be quite sensible to just stop developing these things any further, but i think it’s completely naïve to think that would happen.

there’s no way to make that happen.

slimboyfat

Okay notsurewheretogo, I will try to listen to that video, but it is worth noting that DeGrasse Tyson has no discernible expertise in AI. Hinton, in the original video is a leading expert on the topic. That makes a huge difference. The pattern seems to be that the greater expertise the person has in AI, the more worried they are about it.

DeGrasse Tyson is a public intellectual who unfortunately people turn to for opinions on all sorts of things from lab leak theories, to vaccine safety, and political issues. He will quite happily give his views on any topic. It doesn’t mean he knows what he’s talking about. He is an expert on astronomy and physics and used to concentrate mainly on issues such whether Pluto is a planet or not. That’s his area of expertise.

I’m not saying there won’t be any value in the video, but it doesn’t have anywhere near the same weight, or it shouldn’t have, as input from Hinton or other experts in the field of AI.

Paul Christiano is an expert in AI who says that it probably won’t kill us. He’s a good person to hear the other side of the argument. But even he thinks the existential risk is real (I think he puts it at about 10% if I remember correctly). He’s more upbeat than some of his colleagues however.

https://youtu.be/GyFkWb903aU