diff --git a/src/ngSanitize/sanitize.js b/src/ngSanitize/sanitize.js
index fae50aac259d..de6dd3dd135d 100644
--- a/src/ngSanitize/sanitize.js
+++ b/src/ngSanitize/sanitize.js
@@ -154,11 +154,11 @@ function sanitizeText(chars) {
 
 // Regular Expressions for parsing tags and attributes
 var START_TAG_REGEXP =
-       /^<\s*([\w:-]+)((?:\s+[\w:-]+(?:\s*=\s*(?:(?:"[^"]*")|(?:'[^']*')|[^>\s]+))?)*)\s*(\/?)\s*>/,
-  END_TAG_REGEXP = /^<\s*\/\s*([\w:-]+)[^>]*>/,
+       /^<((?:[a-zA-Z])[\w:-]*)((?:\s+[\w:-]+(?:\s*=\s*(?:(?:"[^"]*")|(?:'[^']*')|[^>\s]+))?)*)\s*(\/?)\s*(>?)/,
+  END_TAG_REGEXP = /^<\/\s*([\w:-]+)[^>]*>/,
   ATTR_REGEXP = /([\w:-]+)(?:\s*=\s*(?:(?:"((?:[^"])*)")|(?:'((?:[^'])*)')|([^>\s]+)))?/g,
   BEGIN_TAG_REGEXP = /^</,
-  BEGING_END_TAGE_REGEXP = /^<\s*\//,
+  BEGING_END_TAGE_REGEXP = /^<\//,
   COMMENT_REGEXP = /<!--(.*?)-->/g,
   DOCTYPE_REGEXP = /<!DOCTYPE([^>]*?)>/i,
   CDATA_REGEXP = /<!\[CDATA\[(.*?)]]>/g,
@@ -232,10 +232,11 @@ function makeMap(str) {
  * @param {object} handler
  */
 function htmlParser( html, handler ) {
-  var index, chars, match, stack = [], last = html;
+  var index, chars, match, stack = [], last = html, text;
   stack.last = function() { return stack[ stack.length - 1 ]; };
 
   while ( html ) {
+    text = '';
     chars = true;
 
     // Make sure we're not in a script or style element
@@ -274,16 +275,23 @@ function htmlParser( html, handler ) {
         match = html.match( START_TAG_REGEXP );
 
         if ( match ) {
-          html = html.substring( match[0].length );
-          match[0].replace( START_TAG_REGEXP, parseStartTag );
+          // We only have a valid start-tag if there is a '>'.
+          if ( match[4] ) {
+            html = html.substring( match[0].length );
+            match[0].replace( START_TAG_REGEXP, parseStartTag );
+          }
           chars = false;
+        } else {
+          // no ending tag found --- this piece should be encoded as an entity.
+          text += '<';
+          html = html.substring(1);
         }
       }
 
       if ( chars ) {
         index = html.indexOf("<");
 
-        var text = index < 0 ? html : html.substring( 0, index );
+        text += index < 0 ? html : html.substring( 0, index );
         html = index < 0 ? "" : html.substring( index );
 
         if (handler.chars) handler.chars( decodeEntities(text) );
diff --git a/test/ngSanitize/sanitizeSpec.js b/test/ngSanitize/sanitizeSpec.js
index 76a48087576e..5f2594397657 100644
--- a/test/ngSanitize/sanitizeSpec.js
+++ b/test/ngSanitize/sanitizeSpec.js
@@ -21,6 +21,7 @@ describe('HTML', function() {
 
     var handler, start, text, comment;
     beforeEach(function() {
+      text = "";
       handler = {
         start: function(tag, attrs, unary){
           start = {
@@ -35,7 +36,7 @@ describe('HTML', function() {
           });
         },
         chars: function(text_){
-          text = text_;
+          text += text_;
         },
         end:function(tag) {
           expect(tag).toEqual(start.tag);
@@ -81,8 +82,31 @@ describe('HTML', function() {
       expect(text).toEqual('text');
     });
 
+    it('should not treat "<" followed by a non-/ or non-letter as a tag', function() {
+      expectHTML('<- text1 text2 <1 text1 text2 <{', handler).
+        toBe('&lt;- text1 text2 &lt;1 text1 text2 &lt;{');
+    });
+
+    it('should throw badparse if text content contains "<" followed by "/" without matching ">"', function() {
+      expect(function() {
+        htmlParser('foo </ bar', handler);
+      }).toThrowMinErr('$sanitize', 'badparse', 'The sanitizer was unable to parse the following block of html: </ bar');
+    });
+
+    it('should throw badparse if text content contains "<" followed by an ASCII letter without matching ">"', function() {
+      expect(function() {
+        htmlParser('foo <a bar', handler);
+      }).toThrowMinErr('$sanitize', 'badparse', 'The sanitizer was unable to parse the following block of html: <a bar');
+    });
+
+    it('should accept tag delimiters such as "<" inside real tags', function() {
+      // Assert that the < is part of the text node content, and not part of a tag name.
+      htmlParser('<p> 10 < 100 </p>', handler);
+      expect(text).toEqual(' 10 < 100 ');
+    });
+
     it('should parse newlines in tags', function() {
-      htmlParser('<\ntag\n attr="value"\n>text<\n/\ntag\n>', handler);
+      htmlParser('<tag\n attr="value"\n>text</\ntag\n>', handler);
       expect(start).toEqual({tag:'tag', attrs:{attr:'value'}, unary:false});
       expect(text).toEqual('text');
     });
@@ -123,8 +147,9 @@ describe('HTML', function() {
     expectHTML('a<!DocTyPe html>c.').toEqual('ac.');
   });
 
-  it('should remove nested script', function() {
-    expectHTML('a< SCRIPT >A< SCRIPT >evil< / scrIpt >B< / scrIpt >c.').toEqual('ac.');
+  it('should escape non-start tags', function() {
+    expectHTML('a< SCRIPT >A< SCRIPT >evil< / scrIpt >B< / scrIpt >c.').
+      toBe('a&lt; SCRIPT &gt;A&lt; SCRIPT &gt;evil&lt; / scrIpt &gt;B&lt; / scrIpt &gt;c.');
   });
 
   it('should remove attrs', function() {
@@ -165,14 +190,16 @@ describe('HTML', function() {
     expectHTML(everything).toEqual(everything);
   });
 
-  it('should handle improper html', function() {
+  it('should mangle improper html', function() {
+    // This text is encoded more than a real HTML parser would, but it should render the same.
     expectHTML('< div rel="</div>" alt=abc dir=\'"\' >text< /div>').
-      toEqual('<div rel="&lt;/div&gt;" alt="abc" dir="&#34;">text</div>');
+      toBe('&lt; div rel=&#34;&#34; alt=abc dir=\'&#34;\' &gt;text&lt; /div&gt;');
   });
 
-  it('should handle improper html2', function() {
+  it('should mangle improper html2', function() {
+    // A proper HTML parser would clobber this more in most cases, but it looks reasonable.
     expectHTML('< div rel="</div>" / >').
-      toEqual('<div rel="&lt;/div&gt;"/>');
+      toBe('&lt; div rel=&#34;&#34; / &gt;');
   });
 
   it('should ignore back slash as escape', function() {
@@ -195,6 +222,12 @@ describe('HTML', function() {
     expectHTML('\na\n').toEqual('&#10;a&#10;');
   });
 
+  it('should accept tag delimiters such as "<" inside real tags (with nesting)', function() {
+    //this is an integrated version of the 'should accept tag delimiters such as "<" inside real tags' test
+    expectHTML('<p> 10 < <span>100</span> </p>')
+    .toEqual('<p> 10 &lt; <span>100</span> </p>');
+  });
+
   describe('htmlSanitizerWriter', function() {
     /* global htmlSanitizeWriter: false */
     if (angular.isUndefined(window.htmlSanitizeWriter)) return;